AN EMERGENCY RESPONSE SYSTEM AND METHOD

Technical Field

[0001] The present invention relates to an emergency response system and associated method, a sensing unit for use in the emergency response system, a drone system for use in the emergency response system, a method of controlling the drone system, an embedded electronic device for use in the emergency response system, a method of controlling the embedded electronic device, a fire control device for use in the emergency response system and a driverless fire control vehicle for use in the emergency response system.

Background

[0002] Large scale emergency situations may occur in the form of fires, floods, earthquakes, landslides, typhoons, cyclones etc. Also, smaller scale emergency situations may occur for individuals or groups of individuals in relation to personal injury, becoming lost, potential drowning etc.

[0003] Responding to emergency situations in a time and cost-effective manner can reduce the risk of loss of infrastructure, agriculture, livestock, wildlife as well as human life.

[0004] Much of the current focus in existing processes is on fighting established fires, which are hard to contain and very difficult to extinguish. Usually large fires are only extinguished after they have run out of fuel or the weather changes. Much of the firefighting process is focussed on protecting property and fire containment. In August 2020, California had 6000 lightning strikes in 24 hours, resulting in 560 fires causing major damage.

[0005] There is a desire to improve how emergency situations are detected, monitored, controlled or acted upon.

Summary

[0006] It is an object of the present invention to meet this desire or to substantially overcome, or at least ameliorate, one or more disadvantages of existing emergency response, management and mitigation arrangements.

[0007] Disclosed are arrangements which seek to address the above problems by providing an improved emergency response system and method that effectively makes a determination, or suggests calculated options, on how to respond to the emergency situation based on the analysis of emergency variables using smart interconnected systems.

[0008] The herein described systems and methods are arranged to detect a potentially large number of small fires, where detection is usually difficult, and then utilise resources to put the fires out quickly while they are still small, manageable and so easier to put out.

[0009] For example, the herein described systems and methods may utilise large helicopters to drop small amounts of fire suppressant with pinpoint accuracy on small fires to extinguish them. This may mean that a single helicopter may attend to many fires on a single sortie. The herein described systems and methods may optimise the route using algorithms to increase the number of fires a single helicopter can extinguish on a single sortie. The herein described systems and methods incorporate a layered approach so that, if the resources and response become overloaded, fall back operations can be initiated by the system and method.

[0010] Also disclosed are a sensing unit, a drone system an embedded electronic device, a fire control device and a driverless fire control vehicle, all of which are suitable for use in the emergency response system to assist in analysing the emergency situation and/or responding to the emergency situation.

[0011] The herein described emergency response system may enable fast, directed and effective responses to prevent, minimize and/or effectively respond to emergencies by collecting relevant time-critical and reference materials into one or more servers, analysing the data in real time by the server, for example, using machine learning applications, calculating the likely outcomes of different strategies or scenarios at both macro and micro levels, and provision of information to the relevant response organization (resources) to enable a very fast and directed response to prevent, minimize and/or effectively respond to these emergencies, either at an individual level, local level, state level or national level.

[0012] In a complex emergency situation, such as bushfires for example, rapid decisions made on accurate information are essential, as any delay can see a small fire grow into a large fire. The herein described system replaces guesswork with information delivered by numerous sensors, systems and components in an intelligent decision support system.

[0013] Various novel and innovative systems, methods, devices and concepts are described herein and/or featured in the claims. [0014] For example, an innovative part of the described system and process is the use of multiple observations at different levels of granularity to pinpoint small risks (fires) such that the size of the avoided risk is unrelated to the size of the initial risk. Therefore, the system must err on the side of identifying each and every risk whilst not misclassifying something as a risk when it is not. That is, the multiple observation levels are designed to progressively refine from a large number of false positives (to ensure no false negatives) to a point where there are almost no false positives or false negatives. This may be done through a multiple layered approach using satellites and sensing towers to jointly confirm possible positives, then sensing towers and drones to eliminate false positives. Therefore, the process may pinpoint with certainty by combining satellite infra-red imagery, satellite lightning detection, sensing tower sferic lightning detection and tower imaging detection to create a dense net of positives followed up by a directed drone coverage to eliminate the false positives.

[0015] For example, an innovative part of the described system and process is the targeting of lower risk scenarios with aerial attacks by very controlled targeting of retardant using a variety of means. It has previously been assumed that aerial attacks should be focussed on large risk scenarios and therefore there is no requirement for very accurate delivery, instead resulting in delivery of the entire load of retardant in a general area. The herein described system and process use the intersection of very accurate location data (initially by coordinates and then by a combination of weather and IR/Visual observation and very detailed 3D maps), a requirement to maintain height of airborne resources whilst targeting the fires, e.g. to avoid downdraft accelerating and spreading the fire risk, and a requirement to precisely aim and control the retardant spray/jet. It is preferable to do these three things together and several ways of combining these elements are described herein. For example, the use of retractable steerable sprays that have smart controls, the use of C02 canisters, etc.

[0016] For example, an innovative part of the described system and process is the optimisation of logistics when there are a large number of fire ignitions that must all be dealt with. This involves a layered approach to scheduling that solves the problem of time of flight, size of payload, aerial pattern of ignitions and fuel for the aircraft. The layering may initially calculate the sorties as a pattern of drops and reloads of retardant. The calculation may be updated in real time based on new fire ignitions and the actual deliveries that have altered both fuel load and payload load.

[0017] For example, an innovative part of the described system and process is the combining of sferics (interference) measurements with infrared and visual data so that sensing towers may use triangulation to monitor lightning strikes and determine whether the lightning did strike, the location of the lightning strike, is the system able to confirm that ignition occurred at that location and if not, despatch a drone to the location to investigate further.

[0018] According to a first aspect of the present disclosure, there is provided an emergency response system for responding to an emergency situation, the emergency response system comprising: at least one server; a plurality of sensors in communication with the server via at least one sensor communication channel, wherein the sensors are arranged to sense emergency variables associated with the emergency situation and communicate the emergency variables to the server via the sensor communication channel; wherein the server is arranged to analyse the emergency variables and determine how to respond to the emergency situation based on the analysis of the emergency variables, wherein determining how to respond comprises: i) determining a location where the emergency situation is occurring; ii) determining what kind of emergency situation is occurring; iii) determining a severity level of the emergency situation based on the determined location where the emergency situation is occurring, and the determined kind of emergency situation; and iv) determining which of a plurality of resources to send to respond to the emergency situation based on the determined location where the emergency situation is occurring, the determined kind of emergency situation and the determined severity level, wherein the server is further arranged to generate and output an emergency response to the emergency situation to at least one of the plurality of resources based on the determination of how to respond to the emergency situation.

[0019] According to a second aspect of the present disclosure, there is provided a method of responding to an emergency situation using an emergency response system that comprises at least one server and a plurality of sensors in communication with the server via at least one sensor communication channel, wherein the sensors are arranged to sense emergency variables associated with the emergency situation and communicate the emergency variables to the server via the sensor communication channel, the method comprising the steps of: analysing the emergency variables and determining how to respond to the emergency situation based on the analysis of the emergency variables, wherein determining how to respond comprises: i) determining a location where the emergency situation is occurring; ii) determining what kind of emergency situation is occurring; iii) determining a severity level of the emergency situation based on the determined location where the emergency situation is occurring, and the determined kind of emergency situation; and iv) determining which of a plurality of resources to send to respond to the emergency situation based on the determined location where the emergency situation is occurring, the determined kind of emergency situation and the determined severity level, and generating and outputting an emergency response to the emergency situation to at least one of the plurality of resources based on the determination of how to respond to the emergency situation.

[0020] According to a third aspect of the present disclosure, there is provided a sensing unit for use in an emergency response system, the sensing unit comprising a power source, a communication module and a sensing tower for attaching one or more of a plurality of sensors for sensing emergency variables associated with an emergency situation and communicating the emergency variables to a server via the communication channel.

[0021] According to a fourth aspect of the present disclosure, there is provided a drone system for use in an emergency response system that has at least one server, the drone system comprising: a plurality of drones arranged to communicate with the emergency response system via a drone communication channel, wherein at least one drone comprises at least one drone sensor, wherein the drone sensor is arranged to sense at least one emergency variable associated with the emergency situation and communicate the emergency variable to the server, either directly via the drone communication channel or via more or more further drones to the drone communication channel, to enable the server to analyse the emergency variables and make a determination on how to respond to the emergency situation based on the analysis of the emergency variables.

[0022] According to a fifth aspect of the present disclosure, there is provided a method of controlling a drone system for use in an emergency response system that has at least one server, the method comprising the steps of: communicating, with a plurality of drones, with the emergency response system via a drone communication channel, wherein at least one drone comprises at least one drone sensor, wherein the drone sensor is arranged to sense at least one emergency variable associated with the emergency situation the method further comprising communicating the emergency variable to the server, either directly via the drone communication channel or via one or more further drones to the drone communication channel, to enable the server to analyse the emergency variables and make a determination on how to respond to the emergency situation based on the analysis of the emergency variables.

[0023] According to a sixth aspect of the present disclosure, there is provided an embedded electronic device for use in an emergency response system, the embedded electronic device comprising a processor and a processor readable medium, wherein the processor readable medium has a program recorded thereon, where the program is configured to make the embedded electronic device execute a procedure to communicate with the emergency response system as described herein, wherein the embedded electronic device further comprises one or more of the plurality of sensors arranged to sense emergency variables associated with the emergency situation and communicate the emergency variables to the server via the sensor communication channel.

[0024] According to a seventh aspect of the present disclosure, there is provided a method of controlling an embedded electronic device for use in an emergency control system, the method comprising the steps of the embedded electronic device: communicating with the emergency response system as described herein, wherein the embedded electronic device further comprises one or more of the plurality of sensors arranged to sense emergency variables associated with the emergency situation and the method comprises communicating the emergency variables to the server via the sensor communication channel.

[0025] According to an eighth aspect of the present disclosure, there is provided a fire control device for use in the emergency response system, the fire control device comprising an outer shell and a plurality of frozen carbon dioxide pellets located within the outer shell, wherein the outer shell is arranged to i) insulate the frozen carbon dioxide pellets and ii) deteriorate when heated to allow the plurality of frozen carbon dioxide pellets to disperse.

[0026] According to a ninth aspect of the present disclosure, there is provided a driverless fire control vehicle for use in an emergency response system, the driverless fire control vehicle comprising a body having a heat reflective exterior, at least one engine, a plurality of wheels driven by the engine, a control module, a communication module for communicating with an emergency response system, a fire-retardant release system for releasing a fire-retardant material, wherein the communication module is arranged to receive control signals to enable the control module to control the engine and the fire-retardant release system.

[0027] According to a tenth aspect of the present disclosure, there is provided a lightning ignition prediction system or process comprising an artificial intelligence or machine learning system for predicting the probability of ignition following a lightning strike, wherein the system or process is arranged to determine the probability of ignition based on one or more of: the determined type of lightning in the lightning strike; and the determined location of the lightning strike; wherein the system is further arranged to determine how to respond to the lightning strike based on the determined probability of ignition.

[0028] According to an eleventh aspect of the present disclosure, there is provided a lightning risk analysis system or process comprising an artificial intelligence or machine learning system for analysing risk following a lightning strike, wherein the system or process is arranged to analyse the risk based on one or more of: weather model data; weather conditions at locations where lightning is likely; fuel conditions at locations where lightning is likely; cloud data; and historical data on when and where lightning strikes caused ignition.

[0029] According to a twelfth aspect of the present disclosure, there is provided a lightning ignition risk reduction system or process comprising an artificial intelligence or machine learning system for reducing the number, intensity and frequency of lightning strikes, or causing a lightning strike to hit a safe area, wherein the system predicts when, where and what actions that can be taken to reduce the risk of ignition, with information inputs coming from one or more of: topography, fuel loads etc to prioritize areas of greatest risk; weather model data; data from satellites; data from sensing towers; instruments in the clouds on balloons and/or drones; cloud behaviour model data; experimental data using a cloud behaviour model; and result data based on pre-emptive actions for reducing the intensity and/or frequency of the lightning strikes, with actions to induce safe lightning and/or reduce the number, intensity and frequency of lightning strikes including one or more of: cloud seeding; using an explosion to cause a plasma or ionized column of gas inside the cloud to stabilize the charge within the cloud; creating a plasma between a point on the ground and the thundercloud which may cause the cloud to discharge and create lightning; suspending a wire within a cloud to cause a conductive channel to enable the charge built up in the cloud to discharge through the wire; dropping an object that will fall through down the cloud and produce a column of ionized gas as it falls; and gliding, parachuting or using a rocket or similar means to place a device into the cloud at the right place and at a safe height to cause an ionizing column or a plasma to be created to discharge the cloud.

[0030] Other aspects are also disclosed.

Brief Description of the Drawings

[0031] At least one embodiment of the present invention will now be described with reference to the drawings and appendices, in which:

[0032] Figs. 1 A and 1 B form a schematic block diagram of a general-purpose computer system in the form of a server upon which arrangements described can be practiced;

[0033] Figs. 2A and 2B collectively form a schematic block diagram representation of an electronic device upon which described arrangements can be practised; [0034] Fig. 3 shows an overview of an environment in which an emergency response system may be employed according to an embodiment of the present disclosure;

[0035] Fig. 4 shows a drone control scenario for an emergency response system according to an embodiment of the present disclosure;

[0036] Fig. 5 shows a sensing unit for an emergency response system according to an embodiment of the present disclosure;

[0037] Figs. 6 and 7 show a sensing tower for an emergency response system according to an embodiment of the present disclosure;

[0038] Fig. 8 shows a sensing tower on a power pylon for an emergency response system according to an embodiment of the present disclosure;

[0039] Figs. 9 and 10A show a process for an emergency response system according to an embodiment of the present disclosure;

[0040] Fig. 10B shows a system diagram for an emergency response system according to an embodiment of the present disclosure;

[0041] Fig. 11 shows a fire retardant device for an emergency response system according to an embodiment of the present disclosure;

[0042] Figs. 12 to 14 shows a fire retardant device in use for an emergency response system according to an embodiment of the present disclosure;

[0043] Figs. 15 and 16 show a driverless fire control vehicle for an emergency response system according to an embodiment of the present disclosure.

[0044] Figs. 17A - 17F show a baffle system according to an embodiment of the present disclosure;

[0045] Figs. 18A-18E show a sensing tower according to embodiments of the present disclosure;

[0046] Figs. 19A-19F show a nozzle system according to embodiments of the present disclosure Detailed Description including Best Mode

[0047] Where reference is made in any one or more of the accompanying drawings to steps and/or features, which have the same reference numerals, those steps and/or features have for the purposes of this description the same function(s) or operation(s), unless the contrary intention appears.

[0048] The following embodiment describes an emergency response system in the context of a fire emergency response system. It will be understood that the emergency response system may be used for alternative emergency responses other than fires, and examples of such alternatives are provided herein.

Server Description

[0049] Figs. 1 A and 1 B depict a general-purpose computer system 100 in the form of a server, upon which the various arrangements described herein may be practiced.

[0050] As seen in Fig. 1A, the computer system 100, in the form of a server, includes: a computer module 1301.

[0051] Optionally, the server may have input devices such as a keyboard 1302 and a mouse pointer device 1303, and output devices including a printer 1315, a display device 1314 and loudspeakers 1317. An external Modulator-Demodulator (Modem) transceiver device 1316 may be used by the computer module 1301 for communicating to and from a communications network 1320 via a connection 1321. The communications network 1320 may be a wide-area network (WAN), such as the Internet (305 in Fig. 3), a cellular telecommunications network, or a private WAN. Where the connection 1321 is a telephone line, the modem 1316 may be a traditional “dial-up” modem. Alternatively, where the connection 1321 is a high capacity (e.g., cable) connection, the modem 1316 may be a broadband modem. A wireless modem may also be used for wireless connection to the communications network 1320.

[0052] The computer module 1301 typically includes at least one processor unit 1305, and a memory unit 1306. For example, the memory unit 1306 may have semiconductor random access memory (RAM) and semiconductor read only memory (ROM). The computer module 1301 also includes a number of input/output (I/O) interfaces including: an audio-video interface 1307 that couples to the video display 1314, loudspeakers 1317 and microphone 1380; an I/O interface 1313 that couples to the keyboard 1302, mouse 1303, scanner 1326, camera 1327 and optionally a joystick or other human interface device (not illustrated); and an interface 1308 for the external modem 1316 and printer 1315. In some implementations, the modem 1316 may be incorporated within the computer module 1301, for example within the interface 1308. The computer module 1301 also has a local network interface 1311 , which permits coupling of the computer system 100 via a connection 1323 to a local-area communications network 1322, known as a Local Area Network (LAN). As illustrated in Fig. 1A, the local communications network 1322 may also couple to the wide network 1320 via a connection 1324, which would typically include a so-called “firewall” device or device of similar functionality. The local network interface 1311 may comprise an Ethernet circuit card, a Bluetooth ^® wireless arrangement or an IEEE 802.11 wireless arrangement; however, numerous other types of interfaces may be practiced for the interface 1311.

[0053] The I/O interfaces 1308 and 1313 may afford either or both of serial and parallel connectivity, the former typically being implemented according to the Universal Serial Bus (USB) standards and having corresponding USB connectors (not illustrated). Storage devices 1309 are provided and typically include a hard disk drive (HDD) 1310. Other storage devices such as a floppy disk drive and a magnetic tape drive (not illustrated) may also be used. An optical disk drive 1312 is typically provided to act as a non-volatile source of data. Portable memory devices, such optical disks (e.g., CD-ROM, DVD, Blu-ray Disc™), USB-RAM, portable, external hard drives, and floppy disks, for example, may be used as appropriate sources of data to the system 100.

[0054] The components 1305 to 1313 of the computer module 1301 typically communicate via an interconnected bus 1304 and in a manner that results in a conventional mode of operation of the computer system 100 known to those in the relevant art. For example, the processor 1305 is coupled to the system bus 1304 using a connection 1318. Likewise, the memory 1306 and optical disk drive 1312 are coupled to the system bus 1304 by connections 1319.

[0055] The server methods described herein may be implemented using the computer system 100 wherein the server processes to be described, may be implemented as one or more software application programs 1333 executable within the computer system 100. In particular, the steps of the server processes may be effected by instructions 1331 (see Fig. 1 B) in the software 1333 that are carried out within the computer system 100. The software instructions 1331 may be formed as one or more code modules, each for performing one or more particular tasks. [0056] The software may be stored in a computer readable medium, including the storage devices described below, for example. The software may be loaded into the computer system 100 from the computer readable medium, and then executed by the computer system 100. A computer readable medium having such software or computer program recorded on the computer readable medium is a computer program product. The use of the computer program product in the computer system 100 preferably effects an advantageous apparatus for use in an emergency response system as described herein.

[0057] The software 1333 is typically stored in the HDD 1310 or the memory 1306. The software is loaded into the computer system 100 from a computer readable medium, and executed by the computer system 100. Thus, for example, the software 1333 may be stored on an optically readable disk storage medium (e.g., CD-ROM) 1325 that is read by the optical disk drive 1312. A computer readable medium having such software or computer program recorded on it is a computer program product. The use of the computer program product in the computer system 100 preferably effects an apparatus for use in an emergency response system as described herein.

[0058] In some instances, the application programs 1333 may be supplied to the user encoded on one or more CD-ROMs 1325 and read via the corresponding drive 1312, or alternatively may be read by the user from the networks 1320 or 1322. Still further, the software can also be loaded into the computer system 100 from other computer readable media. Computer readable storage media refers to any non-transitory tangible storage medium that provides recorded instructions and/or data to the computer system 100 for execution and/or processing. Examples of such storage media include floppy disks, magnetic tape, CD-ROM, DVD, Blu-ray™ Disc, a hard disk drive, a ROM or integrated circuit, USB memory, a magneto optical disk, or a computer readable card such as a PCMCIA card and the like, whether or not such devices are internal or external of the computer module 1301. Examples of transitory or non-tangible computer readable transmission media that may also participate in the provision of software, application programs, instructions and/or data to the computer module 1301 include radio or infra-red transmission channels as well as a network connection to another computer or networked device, and the Internet or Intranets including e-mail transmissions and information recorded on Websites and the like.

[0059] The second part of the application programs 1333 and the corresponding code modules mentioned above may be executed to implement one or more graphical user interfaces (GUIs) to be rendered or otherwise represented upon the display 1314. Through manipulation of typically the keyboard 1302 and the mouse 1303, a user of the computer system 100 and the application may manipulate the interface in a functionally adaptable manner to provide controlling commands and/or input to the applications associated with the GUI(s). Other forms of functionally adaptable user interfaces may also be implemented, such as an audio interface utilizing speech prompts output via the loudspeakers 1317 and user voice commands input via the microphone 1380.

[0060] Fig. 1 B is a detailed schematic block diagram of the processor 1305 and a “memory” 1334. The memory 1334 represents a logical aggregation of all the memory modules (including the HDD 1309 and semiconductor memory 1306) that can be accessed by the computer module 1301 in Fig. 1A.

[0061] When the computer module 1301 is initially powered up, a power-on self-test (POST) program 1350 executes. The POST program 1350 is typically stored in a ROM 1349 of the semiconductor memory 1306 of Fig. 1A. A hardware device such as the ROM 1349 storing software is sometimes referred to as firmware. The POST program 1350 examines hardware within the computer module 1301 to ensure proper functioning and typically checks the processor 1305, the memory 1334 (1309, 1306), and a basic input-output systems software (BIOS) module 1351, also typically stored in the ROM 1349, for correct operation. Once the POST program 1350 has run successfully, the BIOS 1351 activates the hard disk drive 1310 of Fig. 1A. Activation of the hard disk drive 1310 causes a bootstrap loader program 1352 that is resident on the hard disk drive 1310 to execute via the processor 1305. This loads an operating system 1353 into the RAM memory 1306, upon which the operating system 1353 commences operation. The operating system 1353 is a system level application, executable by the processor 1305, to fulfil various high level functions, including processor management, memory management, device management, storage management, software application interface, and generic user interface.

[0062] The operating system 1353 manages the memory 1334 (1309, 1306) to ensure that each process or application running on the computer module 1301 has sufficient memory in which to execute without colliding with memory allocated to another process. Furthermore, the different types of memory available in the system 100 of Fig. 1A must be used properly so that each process can run effectively. Accordingly, the aggregated memory 1334 is not intended to illustrate how particular segments of memory are allocated (unless otherwise stated), but rather to provide a general view of the memory accessible by the computer system 100 and how such is used. [0063] As shown in Fig. 1B, the processor 1305 includes a number of functional modules including a control unit 1339, an arithmetic logic unit (ALU) 1340, and a local or internal memory 1348, sometimes called a cache memory. The cache memory 1348 typically includes a number of storage registers 1344 - 1346 in a register section. One or more internal busses 1341 functionally interconnect these functional modules. The processor 1305 typically also has one or more interfaces 1342 for communicating with external devices via the system bus 1304, using a connection 1318. The memory 1334 is coupled to the bus 1304 using a connection 1319.

[0064] The application program 1333 includes a sequence of instructions 1331 that may include conditional branch and loop instructions. The program 1333 may also include data 1332 which is used in execution of the program 1333. The instructions 1331 and the data 1332 are stored in memory locations 1328, 1329, 1330 and 1335, 1336, 1337, respectively. Depending upon the relative size of the instructions 1331 and the memory locations 1328-1330, a particular instruction may be stored in a single memory location as depicted by the instruction shown in the memory location 1330. Alternately, an instruction may be segmented into a number of parts each of which is stored in a separate memory location, as depicted by the instruction segments shown in the memory locations 1328 and 1329.

[0065] In general, the processor 1305 is given a set of instructions which are executed therein. The processor 1305 waits for a subsequent input, to which the processor 1305 reacts to by executing another set of instructions. Each input may be provided from one or more of a number of sources, including data generated by one or more of the input devices 1302, 1303, data received from an external source across one of the networks 1320, 1302, data retrieved from one of the storage devices 1306, 1309 or data retrieved from a storage medium 1325 inserted into the corresponding reader 1312, all depicted in Fig. 1A. The execution of a set of the instructions may in some cases result in output of data. Execution may also involve storing data or variables to the memory 1334.

[0066] The disclosed arrangements use input variables 1354, such as emergency variables for example, which are stored in the memory 1334 in corresponding memory locations 1355, 1356, 1357. The arrangements produce output variables 1361, such as variables associated with an emergency response, which are stored in the memory 1334 in corresponding memory locations 1362, 1363, 1364. Intermediate variables 1358 may be stored in memory locations 1359, 1360, 1366 and 1367. [0067] Referring to the processor 1305 of Fig. 1B, the registers 1344, 1345, 1346, the arithmetic logic unit (ALU) 1340, and the control unit 1339 work together to perform sequences of micro-operations needed to perform “fetch, decode, and execute” cycles for every instruction in the instruction set making up the program 1333. Each fetch, decode, and execute cycle comprises: a fetch operation, which fetches or reads an instruction 1331 from a memory location 1328, 1329, 1330; a decode operation in which the control unit 1339 determines which instruction has been fetched; and an execute operation in which the control unit 1339 and/or the ALU 1340 execute the instruction.

[0068] Thereafter, a further fetch, decode, and execute cycle for the next instruction may be executed. Similarly, a store cycle may be performed by which the control unit 1339 stores or writes a value to a memory location 1332.

[0069] Each step or sub-process described with reference to a server is associated with one or more segments of the program 1333 and is performed by the register section 1344,

1345, 1347, the ALU 1340, and the control unit 1339 in the processor 1305 working together to perform the fetch, decode, and execute cycles for every instruction in the instruction set for the noted segments of the program 1333.

[0070] The server related methods described herein may alternatively be implemented in dedicated hardware such as one or more integrated circuits performing the functions or sub functions of the emergency response system as described. Such dedicated hardware may include graphic processors, digital signal processors, or one or more microprocessors and associated memories.

Embedded Electronic Device Description

[0071] Figs. 2A and 2B collectively form a schematic block diagram of a general purpose electronic device 2001 including embedded components, upon which the emergency response methods to be described may be practiced. The electronic device 2001 may be, for example, a mobile phone, a tablet, a smartphone, or any other suitable device in which processing resources, compared to main computing devices, are limited. Nevertheless, the methods to be described may also be performed on higher-level devices such as desktop computers, server computers, and other such devices with significantly larger processing resources, as discussed above and herein. [0072] As seen in Fig. 2A, the electronic device 2001 comprises an embedded controller 2002. Accordingly, the electronic device 2001 may be referred to as an “embedded device.” In the present example, the controller 2002 has a processing unit (or processor) 2005 which is bi-directionally coupled to an internal storage module 2009. The storage module 2009 may be formed from non-volatile semiconductor read only memory (ROM) 2060 and semiconductor random access memory (RAM) 2070, as seen in Fig. 2B. The RAM 2070 may be volatile, non-volatile or a combination of volatile and non-volatile memory.

[0073] The electronic device 2001 includes a display controller 2007, which is connected to a video display 2014, such as a liquid crystal display (LCD) panel or the like. The display controller 2007 is configured for displaying graphical images on the video display 2014 in accordance with instructions received from the embedded controller 2002, to which the display controller 2007 is connected.

[0074] The electronic device 2001 also includes user input devices 2013 which are typically formed by keys, a keypad or like controls. In some implementations, the user input devices 2013 may include a touch sensitive panel physically associated with the display 2014 to collectively form a touch-screen. Such a touch-screen may thus operate as one form of graphical user interface (GUI) as opposed to a prompt or menu driven GUI typically used with keypad-display combinations. Other forms of user input devices may also be used, such as a microphone (not illustrated) for voice commands or a joystick/thumb wheel (not illustrated) for ease of navigation about menus.

[0075] As seen in Fig. 2A, the electronic device 2001 also comprises a portable memory interface 2006, which is coupled to the processor 2005 via a connection 2019. The portable memory interface 2006 allows a complementary portable memory device 2025 to be coupled to the electronic device 2001 to act as a source or destination of data or to supplement the internal storage module 2009. Examples of such interfaces permit coupling with portable memory devices such as Universal Serial Bus (USB) memory devices, Secure Digital (SD) cards, Personal Computer Memory Card International Association (PCMIA) cards, optical disks and magnetic disks.

[0076] The electronic device 2001 also has a communications interface 2008 to permit coupling of the device 2001 to a computer or communications network such as a wide area network 1320 via a connection 2021. The connection 2021 may be wired or wireless. For example, the connection 2021 may be radio frequency or optical. An example of a wired connection includes Ethernet. Further, an example of wireless connection includes Bluetooth™ type local interconnection, Wi-Fi (including protocols based on the standards of the IEEE 802.11 family), Infrared Data Association (IrDa) and the like.

[0077] Typically, the electronic device 2001 is configured to perform some special function in conjunction with the emergency response system described herein. The embedded controller 2002, possibly in conjunction with further special function components 2010, is provided to perform that special function. The special function components 2010 is connected to the embedded controller 2002 and may represent those components required for communications in a cellular telephone environment.

[0078] Various methods for use by an embedded electronic device in conjunction with the emergency response system described hereinafter may be implemented using the embedded controller 2002, where the various processes described may be implemented as one or more software application (“App”) programs 2033 executable within the embedded controller 2002. The electronic device 2001 of Fig. 2A implements the described methods. In particular, with reference to Fig. 2B, the steps of the described methods are effected by instructions in the software 2033 that are carried out within the controller 2002. The software instructions may be formed as one or more code modules, each for performing one or more particular tasks. The software may also be divided into two separate parts, in which a first part and the corresponding code modules performs the described methods and a second part and the corresponding code modules manage a user interface between the first part and the user.

[0079] The software 2033 of the embedded controller 2002 is typically stored in the non volatile ROM 2060 of the internal storage module 2009. The software 2033 stored in the ROM 2060 can be updated when required from a computer readable medium. The software 2033 can be loaded into and executed by the processor 2005. In some instances, the processor 2005 may execute software instructions that are located in RAM 2070. Software instructions may be loaded into the RAM 2070 by the processor 2005 initiating a copy of one or more code modules from ROM 2060 into RAM 2070. Alternatively, the software instructions of one or more code modules may be pre-installed in a non-volatile region of RAM 2070 by a manufacturer. After one or more code modules have been located in RAM 2070, the processor 2005 may execute software instructions of the one or more code modules.

[0080] The application program 2033 is typically pre-installed and stored in the ROM 2060 by a manufacturer, prior to distribution of the electronic device 2001. However, in some instances, the application programs 2033 may be supplied to the user encoded on one or more CD-ROM (not shown) and read via the portable memory interface 2006 of Fig. 2A prior to storage in the internal storage module 2009 or in the portable memory 2025. In another alternative, the software application program 2033 may be read by the processor 2005 from the network 2020, or loaded into the controller 2002 or the portable storage medium 2025 from other computer readable media. Computer readable storage media refers to any non-transitory tangible storage medium that participates in providing instructions and/or data to the controller 2002 for execution and/or processing. Examples of such storage media include floppy disks, magnetic tape, CD-ROM, a hard disk drive, a ROM or integrated circuit, USB memory, a magneto-optical disk, flash memory, or a computer readable card such as a PCMCIA card and the like, whether or not such devices are internal or external of the device 2001. Examples of transitory or non tangible computer readable transmission media that may also participate in the provision of software, application programs, instructions and/or data to the device 2001 include radio or infra-red transmission channels as well as a network connection to another computer or networked device, and the Internet or Intranets including e-mail transmissions and information recorded on Websites and the like. A computer readable medium having such software or computer program recorded on it is a computer program product.

[0081] The second part of the application programs 2033 and the corresponding code modules mentioned above may be executed to implement one or more graphical user interfaces (GUIs) to be rendered or otherwise represented upon the display 2014 of Fig. 2A. Through manipulation of the user input device 2013 (e.g., the keypad), a user of the device 2001 and the application programs 2033 may manipulate the interface in a functionally adaptable manner to provide controlling commands and/or input to the applications associated with the GUI(s). Other forms of functionally adaptable user interfaces may also be implemented, such as an audio interface utilizing speech prompts output via loudspeakers (not illustrated) and user voice commands input via the microphone (not illustrated).

[0082] Fig. 2B illustrates in detail the embedded controller 2002 having the processor 2005 for executing the application programs 2033 and the internal storage 2009. The internal storage 2009 comprises read only memory (ROM) 2060 and random access memory (RAM) 2070. The processor 2005 is able to execute the application programs 2033 stored in one or both of the connected memories 2060 and 2070. When the electronic device 2001 is initially powered up, a system program resident in the ROM 2060 is executed. The application program 2033 permanently stored in the ROM 2060 is sometimes referred to as “firmware”. Execution of the firmware by the processor 2005 may fulfil various functions, including processor management, memory management, device management, storage management and user interface. [0083] The processor 2005 typically includes a number of functional modules including a control unit (CU) 2051, an arithmetic logic unit (ALU) 2052, a digital signal processor (DSP)

2053 and a local or internal memory comprising a set of registers 2054 which typically contain atomic data elements 2056, 2057, along with internal buffer or cache memory 2055. One or more internal buses 2059 interconnect these functional modules. The processor 2005 typically also has one or more interfaces 2058 for communicating with external devices via system bus 2081, using a connection 2061.

[0084] The application program 2033 includes a sequence of instructions 2062 through 2063 that may include conditional branch and loop instructions. The program 2033 may also include data, which is used in execution of the program 2033. This data may be stored as part of the instruction or in a separate location 2064 within the ROM 2060 or RAM 2070.

[0085] In general, the processor 2005 is given a set of instructions, which are executed therein. This set of instructions may be organised into blocks, which perform specific tasks or handle specific events that occur in the electronic device 2001. Typically, the application program 2033 waits for events and subsequently executes the block of code associated with that event. Events may be triggered in response to input from a user, via the user input devices 2013 of Fig. 2A, as detected by the processor 2005. Events may also be triggered in response to other sensors and interfaces in the electronic device 2001.

[0086] The execution of a set of the instructions may require numeric variables to be read and modified. Such numeric variables are stored in the RAM 2070. The disclosed method uses input variables 2071 that are stored in known locations 2072, 2073 in the memory 2070. The input variables 2071 are processed to produce output variables 2077 that are stored in known locations 2078, 2079 in the memory 2070. Intermediate variables 2074 may be stored in additional memory locations in locations 2075, 2076 of the memory 2070. Alternatively, some intermediate variables may only exist in the registers 2054 of the processor 2005.

[0087] The execution of a sequence of instructions is achieved in the processor 2005 by repeated application of a fetch-execute cycle. The control unit 2051 of the processor 2005 maintains a register called the program counter, which contains the address in ROM 2060 or RAM 2070 of the next instruction to be executed. At the start of the fetch execute cycle, the contents of the memory address indexed by the program counter is loaded into the control unit 2051. The instruction thus loaded controls the subsequent operation of the processor 2005, causing for example, data to be loaded from ROM memory 2060 into processor registers 2054, the contents of a register to be arithmetically combined with the contents of another register, the contents of a register to be written to the location stored in another register and so on. At the end of the fetch execute cycle the program counter is updated to point to the next instruction in the system program code. Depending on the instruction just executed this may involve incrementing the address contained in the program counter or loading the program counter with a new address in order to achieve a branch operation.

[0088] Each step or sub-process in the processes of the methods described below is associated with one or more segments of the application program 2033, and is performed by repeated execution of a fetch-execute cycle in the processor 2005 or similar programmatic operation of other independent processor blocks in the electronic device 2001.

[0089] The electronic device 2001 also has one or more sensors 307 for sensing emergency variables as described herein. For example, the sensor(s) in the electronic device may be one or more of a heat sensor, a magnetic field sensor, a visual sensor, an audio sensor, a location sensor, and an orientation sensor, for example.

General Environment Description

[0090] Referring to Fig. 3, an example environment 301 is depicted for deploying an emergency response system.

[0091] An example emergency response system in the form of a fire response system is depicted in Fig. 3. The fire response system is used to respond to an emergency situation such as an evolving bushfire, for example.

[0092] At least one server 100 is provided in a building 303. The server is in the form of a computer as described above with reference to Figs 1A and 1 B. In this example, the server receives communications and communicates via a wide area network, such as the Internet as depicted by “clouds” 305 in Fig. 3. The server may also receive communications and communicate via one or more alternative communication media as described herein.

[0093] Sensors 307 are depicted in Fig. 3 as a box with a X inside. These sensors are configured or arranged to be in communication with the server via one or more sensor communication channels. In this example, the sensors communicate with the server either directly via the Internet, or via alternative communication media such as a satellite communication channel using a satellite 308. [0094] For example, the sensors may communicate with the server via one or more sensor communication channels, such as, a mobile telephone communication channel, a wired telephone communication channel, a co-axial communication channel, an optical fibre communication channel, a power line communication channel, a satellite communication channel, an FM communication channel, an AM communication channel, a line of sight optical communication, and a microwave communication channel. The type of sensor communication channel may be dependent on the type of sensor and the location of the sensor.

[0095] The sensors may be one or more of a number of different types of sensor, such as, for example, a heat sensor, humidity sensor, moisture sensor, rain sensor, smoke sensor, wind vector sensor, wind direction sensor, wind speed or wind velocity sensor, lightning detection sensor, magnetic field sensor, a visual sensor, an audio sensor, a location sensor, and an orientation sensor. For example, one or more wind speed sensors may be adapted or arranged to detect wind velocity at different altitudes. For example, the visual sensor may be one or more cameras arranged to capture images. For example, the audio sensor may be an audio recording device, such as microphone. The location sensor may be a GPS sensor. The orientation sensor may be a gyroscopic sensor. One or more of the sensors that form part of the emergency response system may also be built into commercially available electronic devices, such as mobile phones, tablets and laptops, for example.

[0096] The server may store details of each sensor that has been deployed, along with an associated unique ID for the sensor. The location of the sensor may also be stored by the server. The location may be a fixed location, or may be associated with the location of a movable resource to which the sensor is attached, where the movable resource provides the server with details of its location.

[0097] The sensors are arranged to sense one or more emergency variables associated with the emergency situation and communicate the emergency variables to the server via the sensor communication channel. The emergency variables will be dependent on the type of the sensor being used. That is, the emergency variable may be one or more of a heat variable, humidity variable, moisture variable, rain variable, smoke variable, wind vector variable, wind direction variable, wind speed or velocity variable, lightning detection variable, magnetic field variable, a visual variable, an audio variable, a location variable, and an orientation variable.

[0098] For example, the heat variable may be a temperature reading of the air near the heat sensor. The humidity variable may be a humidity reading of the air near the humidity sensor. The moisture variable may be a moisture reading of the earth, soil, vegetation or air near the moisture sensor. The rain variable may be a rain level variable of the amount of rain that has fallen near the rain sensor over a defined period of time. The smoke variable may be a variable associated with the amount of smoke and/or density of smoke near to the smoke sensor. The wind vector variable may identify wind velocity and wind direction near to the wind vector sensor. The wind direction variable may be a compass direction that identifies the direction the wind is blowing near to the wind direction sensor. The wind speed (or velocity) variable may be a measurement in miles or kilometres per hour (or meters per second) of the wind speed near to the wind speed sensor. The lightning detection variable may be a count and/or location variable associated with lightning strikes that occur near to the lightning detection sensor. The magnetic field variable may be a magnetic field measurement in Gauss or Tesla based on the magnetic field around the magnetic field sensor. The visual variable may be an image taken from a camera sensor where the image can then be analysed by the server. The visual variable may also be an indication that a particular object has been identified near to the visual sensor. The audio variable may be a sound file detected by a recording device such as a microphone. The audio variable may be an indication that a particular sound has been identified near to the audio sensor. The location variable may be a location in the form of a GPS co-ordinate, or physical address. The orientation variable may be a co-ordinate in free space using x, y and z axes and may also include a tilt variable.

[0099] The server is arranged to analyse the emergency variables that have been received from the various sensors, and determine how to respond to the emergency situation based on that analysis. That is, the processor in the server acts upon the software instructions incorporated into the software program that runs the server to analyse the emergency variables.

[00100] Various non-limiting examples of different scenarios of how the server analyses the emergency variables and makes a determination on how to respond to the emergency situation are provided herein.

[00101] In general, the server is arranged to make the determination of how to respond by i) determining a location where the emergency situation is occurring; ii) determining what kind of emergency situation is occurring; iii) determining a severity level of the emergency situation based on the determined location where the emergency situation is occurring, and the determined kind of emergency situation; and iv) determining which and how many of a multitude of resources to send to respond to the emergency situation based on the determined location where the emergency situation is occurring, the determined kind of emergency situation and the determined severity level. [00102] For example, the location of the emergency situation can be determined from one or more sensors such as, for example, heat sensors, smoke sensors, location sensors, visual sensors, audio sensors or lightning detection sensors. That is, the emergency variables provided by these sensors can assist in the server determining the location of the start or continuing development of an emergency situation.

[00103] For example, the server may determine which of a kind of emergency situation (i.e. what type of emergency situation) is occurring can be determined by the server by analysing the emergency variables received from the sensors and comparing those variables with known scenarios to see if the received variables match the variables for known scenarios. For example, the kind of emergency situation may be a possible or developing bush fire, where the emergency variable is received from one or more sensors, such as for example, heat sensors, smoke sensors, one or more cameras in the form of visual sensors, one or more microphones in the form of audio sensors, and lightning detection sensors. For a heat sensor, if the emergency variable (the measured heat value) matches or is within a defined threshold, the server may determine that a fire has started or is about to start. Likewise, for a smoke sensor, if the emergency variable (the density of smoke value) matches or is within a defined threshold, the server may determine that a fire has started or is about to start. For a camera sensor, images may be analysed to determine whether smoke, flames or lightning is occurring such that the server may determine that a fire has started or is about to start. For a microphone, sound files may be analysed to determine whether the sounds relate to noise associated with fire (e.g. cracking sounds or animal noises) or such that the server may determine that a fire has started or is about to start. For a lightning detection sensor, the output of the sensor causes the server to determine that lightning has probably struck in a certain location (within a threshold) and that this may (based on one or more other emergency variables and known scenarios) result in the start of a fire.

[00104] Described herein is a lightning ignition prediction system or process that utilises an Artificial Intelligence or machine learning system as described herein to predict the probability of an ignition strike. The lightning ignition prediction system may be used to schedule responses to lightning strikes with inputs including the type of lightning, such as dry lightning or lightning with rain, negative lightning, positive lightning, continuous current lightning. Another input may be the determined location of the lightning strike. A further input may be weather condition data at the determined location. A further input may be the determined fuel load and state at the determined location. A further input may be the topography of the determined location. A further input may be determined location of other lightning strikes to schedule an efficient route for an aerial resource to respond to multiple lightning strikes in the one sortie. A further input may be results of other similar lightning strikes, e.g. whether those ignitions start ignitions. A further input may be estimates of the cost of not responding to particular lightning strikes.

[00105] That is, described herein is a lightning ignition prediction system or process that utilises an artificial intelligence or machine learning system for predicting the probability of ignition following a lightning strike. The system is arranged to determine the probability of ignition based on one or more of: the determined type of lightning in the lightning strike; and the determined location of the lightning strike. The system is also arranged to determine how to respond to the lightning strike based on the determined probability of ignition.

[00106] Also, the lightning ignition prediction system or process may be arranged to respond by scheduling a response by utilising one or more resources of an emergency response system.

[00107] Further, the lightning ignition prediction system or process may be arranged to determine how to respond based on one or more of i) locations of other lightning strikes, ii) whether other lightning strikes resulted in an ignition following the other lightning strikes, and iii) cost estimates associated with not responding to the lightning strike.

[00108] Described herein is a lightning risk system or process that is used to pre-emptively relocate aerial resources to a likely dry lightning storm with inputs from weather models, weather and fuel conditions at locations where lightning is likely, information from satellites, towers and drones about the cloud conditions, and historical information about when and where lightning strikes ignited fires

[00109] That is, described herein is a lightning risk analysis system or process that utilises an artificial intelligence or machine learning system for analysing risk following a lightning strike. The system is arranged to analyse the risk based on one or more of: weather model data, weather conditions at locations where lightning is likely, fuel conditions at locations where lightning is likely, cloud data; and historical data on when and where lightning strikes caused ignition.

[00110] Described herein is a lightning risk reduction system or process that utilises a machine learning system to predict what, when and where action could be taken to reduce the intensity and frequency of lightning strikes with inputs from weather models, measurements from satellites and/or sensing towers, cloud behaviour models, virtual experiments conducted using the cloud behaviour models and results of actual pre-emptive actions to reduce the intensity and frequency of lightning. [00111] That is, described herein is a lightning ignition risk reduction system or process that utilises an artificial intelligence or machine learning system for reducing the number, intensity and frequency of lightning strikes, or causing a lightning strike to hit a safe area. The system is arranged to predict when, where and what actions that can be taken to reduce the risk of ignition. Information inputs to the system come from one or more of: topography, fuel loads etc to prioritize areas of greatest risk, weather model data, data from satellites, data from sensing towers, instruments in the clouds on balloons and/or drones, cloud behaviour model data, experimental data using a cloud behaviour model, result data based on pre-emptive actions for reducing the intensity and/or frequency of the lightning strikes. The system or process includes actions to induce safe lightning and/or reduce the number, intensity and frequency of lightning strikes including one or more of: cloud seeding, using an explosion to cause a plasma or ionized column of gas inside the cloud to stabilize the charge within the cloud, creating a plasma between a point on the ground and the thundercloud which may cause the cloud to discharge and create lightning, suspending a wire within a cloud to cause a conductive channel to enable the charge built up in the cloud to discharge through the wire, dropping an object that will fall through down the cloud and produce a column of ionized gas as it falls, and gliding, parachuting or using a rocket or similar means to place a device into the cloud at the right place and at a safe height to cause an ionizing column or a plasma to be created to discharge the cloud.

[00112] The lightning ignition risk reduction system or process may use data from the satellites that defines one or more of: cloud shape, colour and opacity, area, height and volume of the cloud, speed and direction of the cloud movement, the distribution of water within the cloud.

[00113] The lightning ignition risk reduction system or process may use data from the sensing towers that defines one or more of: cloud height above the ground, cloud colour and opacity, direction and speed of cloud movement, charge in the air.

[00114] The lightning ignition risk reduction system or process may use result data that is based on modelling of clouds in the cloud behaviour model, actual responses to reduce the number, intensity and frequency of lightning strikes, or causing a lightning strike to hit a safe area.

[00115] The server may determine a severity level of the emergency situation based on the determined location where the emergency situation is occurring, and the determined kind of emergency situation. The server, may, for example, determine the severity level based on the type of the fire such as undergrowth, tree canopy, the vegetation proximate to the fire, the temperature and size of the fire, estimates or measurement of the time the fire has been alight, estimates or measurements of the rate of growth of the fire, the weather variables such as temperature, humidity, wind and wind direction, the infrastructure at risk, including proximity to vulnerable infrastructure.

[00116] For example, the server may determine which and how many of multiple resources to send to respond to the emergency situation based on the determined location where the emergency situation is occurring, the determined kind of emergency situation and the determined severity level.

[00117] Resources may include, for example, one or more of aeroplanes 309, manned helicopters 311, unmanned helicopters 313, unmanned aerial vehicles (UAVs), drones 315, manned vehicles, automated unmanned vehicles 317, emergency workers.

[00118] For example, the resources may be used to verify the emergency situation, monitor the emergency situation or mitigate the emergency situation. For example, the resources may be used to apply fire retardant materials in an attempt to control the fire.

[00119] Various resources are described in more detail herein.

[00120] The server is further arranged to generate and output an emergency response to the emergency situation to at least one of the resources based on the determination of how to respond to the emergency situation.

[00121] The server may also be arranged to monitor the emergency situation following the output of the emergency response. That is, the server may determine how effective the emergency response is based at least on the sensed emergency variables. For example, the currently sensed emergency variables may be compared to the previously sensed emergency variables (as discussed above) to determine whether a sufficient change (e.g. based on a threshold) has occurred in the emergency variables to determine an effectiveness value for the control of the emergency situation.

[00122] The server may also be arranged to determine whether and/or how to adapt the emergency response based on the evaluation of the effectiveness of the emergency response. For example, if the effectiveness value is below a defined threshold, the server may determine that one or more further or additional resources are required and so send out a new or modified emergency response. [00123] As a further example, the data and variables associated with a particular emergency response and emergency situation may be stored. This data may be analysed by a machine learning or artificial intelligence (Al) system to assess and adapt emergency responses for future emergency situations based on the analysis. The machine learning and/or Al may be carried out by the server or an alternative computing system in communication with the server. The adaptation of the emergency response may be performed in real time based on data and emergency variables that are fed back to the server and/or to the machine learning/AI system.

[00124] The server may also be arranged to receive environmental data. For example, this environmental data may be data associated with one or more of i) weather data, ii) terrain data, iii) infrastructure data, iv) vulnerable infrastructure data, v) vegetal location data, vi) stock and wildlife location data, and vii) historic lightning strike data.

[00125] For example, weather data may include one or more sources of data. Weather data may be obtained from a Bureau of Meteorology (BOM) database 319 via any suitable communication means as described herein. More specific weather data can be obtained from fixed sensors and from sensors on planes, helicopters, drones, cars and trucks. Additional information related to the weather data can be provided by members of the public using a specially designed emergency App, as described herein.

[00126] For example, terrain data may include one or more sources of information such as mapping databases, including national mapping databases and privately owned databases maintained by organizations such as miners, utilities, road, rail and telecommunications network operators, federal and state government departments and farmers, aerial photography, and satellite imagery. For example, terrain data may be obtained from a Geographic Information Systems (GIS) database 321 via any suitable communication means as described herein. Additional information related to terrain data can be provided by members of the public using a specially designed emergency App, as described herein.

[00127] For example, infrastructure data may include one or more sources including Geographic Information Systems, land ownership records databases, databases owned by electricity, gas and water utilities, databases owned by rail, road and telecommunications network operators, mapping databases including national mapping databases, aerial photographs, satellite information including information which is made available by commercial data collection and analysis organizations, data from public and private owners, and information related to infrastructure data supplied by members of the public using a specially designed emergency App, as described herein. [00128] For example, vulnerable infrastructure data may include one or more sources including mapping databases showing vegetation, data about proximate vegetation held by infrastructure owners such as utilities, road and rail network operators and owners of private infrastructure, and information related to vulnerable infrastructure data provided by members of the public using a specially designed emergency App, as described herein.

[00129] For example, vegetal location data may include one or more mapping databases showing vegetation, sources such as park management databases, databases from utilities showing nearby vegetation, forest management organizations, tourist and recreational organizations, such a hiking clubs and information related to vegetal location data provided by members of the public using a specially designed emergency App, as described herein.

[00130] For example, stock and wildlife location data may include one or more sources such as information from private stock owners showing the location of stock, aerial photographs linked with recognition software, information provided by people studying animal populations and ecology, and information related to stock and wildlife location data provided by members of the public using a specially designed emergency App, as described herein.

[00131] For example, historic lightning strike data may include one or more sources of information including from a Bureau of Meteorology (BOM) database, data from people studying lightning, especially those who can estimate the probability distribution that lightning will affect certain locations, providers of lightning prediction software, from insurance and coronial records and from the public who remember lightning strike locations.

[00132] The server may then be arranged to make the determination of how to respond to the emergency situation based on a combination of the analysis of the emergency variables and a further analysis of the environmental data.

[00133] The server may also have a prioritisation module and a scheduling module formed therein or in communication with the server.

[00134] The prioritisation module may be arranged to determine a priority ranking for one or more of i) infrastructure associated with the infrastructure data or the vulnerable infrastructure data, ii) the location where the emergency situation is occurring, iii) an estimation of the emergency response required, iv) an estimation of a cost associated with not responding to the emergency response. [00135] The scheduling module may be arranged to schedule the resources being sent to respond to the emergency situation based on the priority ranking. For example, a potential fire can be responded to be sending almost any resource equipped with suitable sensors to ascertain whether the potential fire is an actual fire that needs suppressing or is not a fire. A small fire can be suppressed or extinguished by a large number of resources, from land based resources including individual humans, provided they can reach the fire quickly, to drones, small planes, helicopters (both manned and unmanned) and larger water bombers. Substantial fires can only be addressed by large bombers releasing fire suppressants. There is little value sending a resource to a fire if that resource will likely be ineffective. For example, a small fire detected in a forest close to a town that can quickly spread if not responded to will be higher priority that a small fire in an inaccessible valley that is not close to infrastructure and is likely to be contained by a cliff wall and so unlikely to spread. If there are not enough resources to respond to both fires, the fire near the town would have a higher priority than the other forest fire.

[00136] The emergency response system may also have one or more drones 315. The drones may have at least one sensor located thereon. The sensor may be any suitable sensor as described herein. The drone may be arranged to communicate with the server directly through a drone communication channel that is in communication with the server. Alternatively, the drone may be arranged to communicate with the server via at least one further drone that is connected to the drone communication channel that is in communication with the server. For example, multiple drones may communicate with each other via Wi-Fi or any other suitable communication medium.

[00137] A mesh network 323 of interconnected drones may be deployed.

[00138] For example, the drone communication channel may be one or a combination of a mobile telephone communication channel, a satellite communication channel, an FM communication channel, an AM communication channel, a microwave communication channel, a Wi-Fi communication channel, an optical communication channel.

[00139] Fig. 4 shows an example scenario of two drones (315A, 315B) under control via a satellite 308 using a satellite communication channel 401. The drones in this example are controlled to fly underneath the canopy 403 of trees to monitor the environment below the canopy including the undergrowth 405. Sensors 307 onboard the drones monitor the environment and are used to sense emergency variables as described herein. For example, the sensors may include a smoke sensor, a fire sensor etc. The drones may also have a further communication device 407 to enable the drones to communicate with each other. For example, the communication device 407 may use Wi-Fi or any other suitable communication means to enable the drones to communicate with each other. The drones also have one or more propellers 409. The drones may also have an embedded electronic device 411 (e.g. a mobile phone) placed therein, where the device 411 operates as described herein with reference to the device shown and described with reference to Figs. 2A and 2B.

[00140] One or more drone systems are provided that may be used with the emergency response system as described herein.

[00141] The drone system may have multiple drones (as depicted in Fig. 3) arranged to communicate with the emergency response system via a drone communication channel.

[00142] One or more of the drones may have one or more drone sensors, where the drone sensors are any suitable sensor as described herein. For example, the drone sensors may be one or more of a heat sensor, humidity sensor, moisture sensor, smoke sensor, wind direction sensor, wind speed sensor, lightning detection sensor, a visual sensor, an audio sensor, a location sensor, an orientation sensor, and a magnetic field sensor.

[00143] Each drone sensor is arranged to sense at least one emergency variable associated with the emergency situation and communicate the emergency variable to the server via any suitable communication protocol. The communication to the server may be direct via the drone communication channel. Alternatively, the communication to the server may be via one or more further drones using any suitable communication method, such as Wi-Fi for example. The emergency variables are used by the server as described herein to analyse the emergency variables and make a determination on how to respond to the emergency situation based on the analysis of the emergency variables.

[00144] The drone system may have multiple drones that are arranged to communicate with each other via a mesh communication network using any suitable communication protocol, such as Wi-Fi, Bluetooth for example.

[00145] One or more of the drones in the drone system may have a fire-retardant release system for releasing a fire-retardant material or system as described herein. For example, the release system may be a controllable catch to release the fire-retardant material, where the signal to release the catch is generated by the drone, or generated externally (e.g. by the server) and then communicated to the drone. [00146] The emergency response system may also have one or more sensing units 501 as depicted, for example, in Fig. 5. The sensing unit may have a power source, which may be in the form of one or more solar power sources 503 with batteries for providing power to the sensing unit. Alternative power sources may be used as an alternative or together with the solar power source. For example, a battery power source, a generator power source or a connection to the power grid may be provided.

[00147] The sensing unit may also have a sensing tower 504 with a cone base positioned on the sensing unit platform.

[00148] The sensing unit may also have a communication module 505 for communicating via the sensor communication channel with the server or with other components of the emergency response system. The communication module may have a satellite communication system 507 and/or a mobile telephone and/or a Wi-Fi communication system 509

[00149] On the sensing tower are attached multiple sensors as described below. A lifting point 512 may be placed at the top of the sensing tower to enable the tower to be transported as discussed herein. For example, the sensing tower may be moved from its current location to another location, e.g. during winter months, when there is no need for a fire sensor at the current location.

[00150] The sensing unit may also have a GPS unit (a location sensor) for determining a location of the sensing unit in a sensor box 511 along with any other suitable sensors.

[00151] It will be understood that the sensors may include any suitable type of sensor such as, for example, one or more of a heat sensor, humidity sensor, moisture sensor, smoke sensor, wind direction sensor, wind speed sensor, lightning detection sensor, a visual sensor, an audio sensor, a location sensor, an orientation sensor, and magnetic field sensor.

[00152] Where the sensing unit is being used as a remote sensing unit for sensing emergency variables in remote environment, the sensing unit may also have a drone landing pad 513 that has a drone landing surface with a drone recharging interface 514 formed therein to connect with an extendable robotic recharging arm 516 on a drone 315. The drone recharging interface is powered by the power source and used to charge drones 315 that communicate with the communication module 505 to locate the sensing unit utilising the GPS unit, for example. As an alternative to the GPS unit, the location of the sensing unit may be stored in the server. A unique ID associated with the sensing unit may also be stored in the server. [00153] On the sensing tower 504 are positioned one or more sets (515A, 515B) of visual sensors 517 on or as part of a visual sensor platform (519A, 519B). In this example, the visual sensors are multiple cameras that have a wide field of view.

[00154] According to one example, there may be at least two visual sensor platforms, where each platform has multiple visual sensors (e.g. cameras) that are attached to the platform. A first of the visual sensor platforms is located above a second of the visual sensor platforms. Audio sensors may also be incorporated into one or both of these platforms (or anywhere else on the sensing tower).

[00155] The first and second visual sensor platforms are arranged to sense and develop the emergency variables in the form of visual emergency variables based on stereoscopic vision data captured by the visual sensors. The visual emergency variables are then transmitted via the communication module to the server (either directly or indirectly). The visual emergency variables are then analysed by the server to detect the emergency situation, and also to detect how far the emergency situation is occurring from the location of the sensing unit based on the stereoscopic vision data, as will be understood by a person skilled in the art. Stereoscopic cameras mounted on aerial and/or land vehicles may provide real time 3D information about the state of roads back to the server and a look forward service may be provided where drivers can be informed by the server about e.g. road conditions around a corner that they cannot see from their present location. This may be used to reduce accidents due to e.g. bad road conditions, such as landslides, road subsidence, flooded roads, snow or ice on the roads.

[00156] According to one example, shown in Fig. 6, the visual sensors 517 may be arranged on and around platforms (519A, 519B) having an octagonal shape, where on each of the eight sides of the octagonal platform, two visual sensors are positioned. A first is positioned at one distal end of their respective side, while a second is positioned at the opposing distal end of the respective side. Also, in this example, the two platforms are offset from each other rotationally by 22.5 degrees. This arrangement enables the two visual sensors to detect an object 601 in the emergency situation and calculate the distance the object is from the visual sensors. This data may be communicated to the server as part of the emergency variables.

[00157] In Fig. 7, an example sensing tower 504 without a sensing unit is depicted. It will be understood that a sensing unit 501 may be used as an alternative to the sensing tower 504.

The sensing tower in this scenario has its own power source built in. In this example the sensing tower is located on a cliff 701 to enable the visual sensors to view the surrounding environment and detect objects 601. [00158] Fig. 8 shows an example scenario with a power pylon 801 having power lines 803 being used as part of the emergency response system. The pylon may have attached thereto one or more suitable sensors 307, such as, for example, a smoke sensor 307A for detecting emergency variables as described herein.

[00159] An example sensing tower 504 without a sensing unit is depicted. It will be understood that a sensing unit 501 may be used as an alternative to the sensing tower 504. The sensing tower in this scenario has its own power source built in, or obtains power directly from the power pylon 801 via a transformer. The power grid may also be used as a further communication media to communicate with the server or other parts of the emergency response system.

[00160] Fig. 9 shows an example process flow diagram in accordance with the emergency response system as described herein.

[00161] The process starts at step S901. At step S903, the process senses emergency variables associated with the emergency situation, as described herein. At step S905, the process communicates the emergency variables to the server, as described herein. At step S907, the process analyses the emergency variables, as described herein. At step S909, the process determines how to respond to the emergency situation, as described herein.

[00162] The process to determine how to respond to the emergency situation includes the following steps. At step S911, the process determines the location of the emergency situation, as described herein. At step S913, the process determines the kind of emergency situation that is occurring, as described herein. At step S915, the process determines the severity level of the emergency situation, as described herein. At step S917, the process determines the resources to send to the emergency situation, as described herein.

[00163] The process then continues at step s919, where the process outputs the determination to the resources, as described herein. The process then ends at step S921.

[00164] Fig. 10A shows a further example of process flow diagram in accordance with the emergency response system as described herein.

[00165] The process starts at step S1001. At step S1003, the process senses emergency variables associated with the emergency situation, as described herein. At step S1005, the process communicates the emergency variables to the server, as described herein. At step S1007, the process analyses the emergency variables, as described herein. At step S1009, the process determines how to respond to the emergency situation, as described herein.

[00166] The process to determine how to respond to the emergency situation includes the following steps. At step S1011, the process determines the location of the emergency situation, as described herein. At step S1013, the process determines the kind of emergency situation that is occurring, as described herein. At step S1015, the process determines the severity level of the emergency situation, as described herein. At step S1017, the process determines the resources to send to the emergency situation, as described herein. At step S1019, the process outputs the determination to the resources, as described herein.

[00167] The process then continues at step S1021, where the process monitors the emergency situation following the output of the emergency response, as described herein. At step S1023, the process determines how effective the emergency response is based at least on the sensed emergency variables. At step S1025, the process determines whether to adapt the emergency response based on the effectiveness evaluation. If the process determines at step S1025 that the emergency response should be adapted, the process moves to step S1009 to further determine how to respond based on steps S1011, S1013, S1015 and S1017 using the current emergency variables.

[00168] If the process determines at step S1025 that the emergency response should not be adapted, the process ends at step S1027.

[00169] Fig. 10B shows a system and process diagram in accordance with a further example. Existing database information 1001 is provided to the emergency response system 1003 that includes a prioritisation module 1005 and a scheduling module 1007. The existing databases may include any number of different datasets including, for example, terrain data, vegetation data, weather (BOM) data, location of infrastructure data, value of infrastructure data, vulnerability of infrastructure data, stock and wildlife data, lightning strike history data (including location data), simulation data for rates of spread of fires.

[00170] The system also includes real time data obtained from one or more resources 1009, such as, for example, sensing towers and sensors, drones and sensors, embedded electronic device software (Apps), aircraft/helicopters etc. to provide accurate emergency variables to the prioritisation module. [00171] The location of confirmed fires by size and type can be determined 1011 and provided to the prioritisation module by the system.

[00172] Analysis 1013 of locations of possible lightning strikes and other potential fires can be performed by the system.

[00173] Response objectives 1015 can be output or input into the system.

[00174] A database 1017 of the type and capacity of the response resources may be provided to the scheduling module 1007.

[00175] The exact location and readiness status 1019 of the response resources in real time may be provided to the scheduling module 1007.

[00176] One or more emergency responses may be output 1021 by the scheduling module 1007.

[00177] Post response surveillance is performed 1023 to assess the effectiveness of the response(s).

[00178] As shown in Fig. 3 one or more embedded electronic devices (e.g. mobile telephones, smartphones, tablets etc.) 325 in the form as described with reference to the electronic device 2001 described in Figs 2A and 2B may form part of the emergency response system.

[00179] The embedded electronic device software, or App, may be used to operate the electronic device in a specific way as part of the emergency response system.

[00180] For example, the program in the electronic device may be configured to make the embedded electronic device execute a procedure to communicate with the emergency response system via any suitable communication method or protocol.

[00181] The embedded electronic device may have one or more sensors that are arranged to sense emergency variables associated with the emergency situation and the program is arranged to cause the electronic device to communicate the emergency variables to the server via the sensor communication channel, either directly or via other parts of the emergency response system. [00182] The program may also enable the embedded electronic device to register with the server to become a registered device. Device location data associated with the registered device may be sent to the server. One or more emergency notifications may be received by the device from the server upon the server determining that the device location data is within a defined distance of the location where an emergency situation is occurring.

[00183] For example, the program may give the server initial emergency variables associated with an emergency situation. The location may be used by the server to assess the relevance of the emergency variables in relation to an already identified emergency. The server may make a determination on whether the emergency variables relate to the existing emergency situation or a new emergency situation, for example a second fire that is close by to an initial fire.

[00184] Also, the program in the electronic device may be configured to communicate the emergency variables to the server upon receiving the emergency notification.

[00185] Fig. 11 shows an example of a fire control device 1101 in the form of a frozen carbon dioxide release system that may be used in the emergency response system. The fire control device has an outer shell 1103 and multiple frozen carbon dioxide pellets 1105 located within the outer shell. The outer shell is arranged to i) insulate the frozen carbon dioxide pellets and ii) deteriorate when heated to allow the frozen carbon dioxide pellets to disperse.

[00186] The outer shell may also have two or more directional elements (1107A, 1107B), such as fins for example, that are attached thereto and positioned relative to the outer shell to cause the fire control device to spin when falling.

[00187] The fire control device may also have a detonation device 1109 and a timer device 1111 that are arranged to detonate the outer shell upon a timer value expiring in the timer device.

[00188] Fig. 12 shows an example of a deployment process for deploying the fire control device of Fig. 11. An aeroplane 309 has an insulated hold 1201 for storing and releasing the fire control devices using a compressed air tank 1203. The release of the fire control devices may be performed in a number of different ways under control of the emergency response system.

[00189] For example, a helicopter and a drone can hover directly over a fire and can drop a fire control device directly on the fire. A manned helicopter may need to hover at a higher and safer altitude than an unmanned helicopter above the fire. The fire control device will explode after it is released to cover the fire in C02 as per 1 in Fig. 14. The likelihood of the fire being extinguished will be increased by dropping additional fire control devices as per 2 and 3 in Fig. 14. The fire control device may not need to spin with a short vertical drop. A helicopter may be able to transport and drop multiple fire control devices on a number of fires. The system may direct the drone or helicopter to the fire and then sensors of the drone or helicopter may then allow the helicopter or drone to position itself directly above the fire. If there is wind detected by a sensor between the height of the drone or helicopter and the ground, and if the velocity of the wind as detected is sufficient to reduce the effectiveness of the fire control device, if the fire control device is dropped directly above the fire, then the system may calculate, based on the sensed wind variables, how far upwind the fire control device(s) should be dropped so that the wind will carry the fire control device and the C02 released from the C02 device onto the fire. The system may calculate the wind correction using the standard Newtonian equation to calculate the time taken for the fire device to fall to the ground, and the upwind distance may be calculated by the time to fall to the ground times the average velocity of the wind in the region between the drone or helicopter and the ground. If the fire control device is released from a plane, then the plane may fly low and slowly above the fire into the wind if there is one. The fire control devices may be ejected at speed in the direction of the tail of the plane to reduce the speed of the fire control devices relative to the fire. The system may calculate the location and timing of the fire control device drop based on Newtonian motion equations for the fire device to fall to the ground, and the upwind distance may be calculated by the time to fall to the ground times the velocity of the fire control device less the average velocity of the wind in the region between the plane and the fire.

[00190] An example description of how to manufacture a fire control device, in the form of a C02 bomb, is now described.

[00191] The aim of the C02 bomb is to create a cloud of C02 directly above and in close proximity to a fire so that the partial pressure of oxygen is sharply removed. The C02 bomb may also absorb energy from the fire but given the energy that can be released by a fire, oxygen starvation, will have the biggest impact in most situations.

[00192] The C02 bombs are made locally, near to the emergency situation, where they are needed. The equipment to make the C02 bombs is containerized and easily transportable on trucks as is a supply of liquid C02 which can either be trucked in or cooled on site. Any suitable electricity supply is utilised to assist in manufacturing the bombs. [00193] To make the C02 dry ice, carbon dioxide gas is first cooled and pressurized to turn it into a liquid. The liquid is then depressurized and allowed to expand back into a gas. This expansion causes a rapid temperature drop, and some of the carbon dioxide freezes into solid pellets of dry ice. Those pellets are then shaped into chunks.

[00194] The larger the bomb, the greater the volume to surface area and the longer the bomb will be able to remain intact. The bombs will be stored in insulated containers whilst in the air prior to being dropped. In most situations, the insulated container will be the bomb dropping device and will be attached to the plane, helicopter and/or drone with bombs and then detached from the plane, helicopter and/or drone when the plane, helicopter and/or drone returns to base, and another insulated container/bomb dropping device is quickly attached to the plane, helicopter and/or drone, allowing a fast turnaround without exposing the bomb to the outside air.

[00195] The bomb can be delivered right over a fire, for example, by a drone that can hover or by a manned or unmanned helicopter. In these examples the bomb can be delivered with pinpoint accuracy and this will significantly increase the effectiveness of the bomb. C02 bombs dropped by helicopters and hovering drones may have a smooth outer surface as they travel a short distance and will not be required to spin.

[00196] As shown in Fig. 13, if the bomb is delivered from an aeroplane then the bomb will be less precisely dropped and several bombs (1101 A, 1101B, 1101C, 1101D) may need to be dropped to ensure that the fire 1301 is suppressed or extinguished. The wind velocity, as measured by sensors in the emergency response system, may be taken into account when calculating when to drop the bombs so that they reach the location of the emergency situation. The bombs may be designed so that those bombs dropped from a plane will spread out when dropped. That is move a small distance along a horizontal path normal to the direction of the plane whilst the bomb is descending towards the fire. This can be achieved by casting in ridges (or fins) that will cause the bomb to spin and move in the air.

[00197] A C02 bomb may be made in many different ways. The bomb has a casing that is thermally insulated. One low cost casing material that may be used is ice, although other suitable lightweight, non-toxic, non-flammable insulating materials may be used.

[00198] The ice shell may be cast in 2 halves. The shell may have a variety of different shapes. One shape may be a sphere. The bottom hemisphere may have a hole cast into the ice shell. The top hemisphere may have the explosive device cast into the ice shell. [00199] The ice may be formed in a hemisphere mould with a smaller hemisphere inserted into the mould. The ice may be removed from the mould by slightly heating the mould to melt a thin film of water. The ice shells may be welded together by having a hot instrument melt the edges of the hemisphere and then the hemispheres are put together on an atmosphere below freezing. When the shell is full of C02 pieces, the weighted bottom of the bomb may be welded onto the shell, and the bomb is complete. The bomb can be stored in a low temperature environment or put straight into the insulated containers used to drop the bombs. The bombs in the insulated containers can be cooled by solid C02.

[00200] The C02 balls can be shaped into different shapes and sizes. Fig. 14 shows an example of three C02 bombs (1, 2, 3) being dropped down onto a fire 1401 in a situation where the wind velocity is low. For example, a drone or helicopter (manned or unmanned) may be used to drop the bombs. The bomb impact on the fire then has an overlapping effect.

[00201] The top of the bomb may be filled with solid C02 pieces which may be cast in multiple shapes. These C02 pieces will typically be larger than the solid C02 pieces in the lower hemisphere because the solid C02 pieces directly above the fire will have a shorter distance to travel to the fire than solid C02 pieces at the top of the bomb which the explosion will cause to travel away from the centre of the fire so that the fire is surrounded by a falling C02 curtain to starve the fire of oxygen.

[00202] Other kinds of bombs are also possible e.g. foam bombs to starve the fire of air.

[00203] A manned helicopter may be used to drop multiple C02 bombs, one at a time, for example.

[00204] According to another example, a large C02 bomb may be constructed from smaller bombs. When dropped from a large cargo plane, for example, a Hercules transporter, the bomb will fall to a pre-set height and then explode. A parachute may be provided to slow the fall of the large bomb. The explosion will scatter the smaller C02 bombs in a pre-set pattern to ensure that as much of the fire as possible will be beneath these smaller bombs. These smaller bombs will subsequently explode and engulf the fire in C02. The bomb may be placed in a single large container, such as a large insulated box that is shaped to maximize the volume of the bomb whilst still conveniently fitting inside the plane. The bomb will be designed for the plane to transport it. For example, a Hercules can lift a 5 tonne bomb which can be made from 100 50kg bombs. The outside of the bomb may be made of any light, insulating non flammable material. An alternative is to construct an ice container. Having the large bomb slowed down by a parachute may assist the ice container to partially melt which reduces the force required to disperse the smaller bombs. The large bomb container is designed so that the smaller C02 bombs will stack on each other so that an optimal distribution will be achieved when the large bomb is exploded. The container with the bombs may be placed on a suitably large aerial vehicle, such as a Hercules transporter, which is then instructed by the system to approach a large fire. The system may provide direction details and drop location details associated with the fire. As a first step, the large container may be dropped from the aerial vehicle at the calculated drop point. As a second step, the individual C02 bombs may be released from the container by an explosion that will disperse the smaller bombs in a predetermined pattern over the fire.

[00205] Further, after a defined time period after release, as a third step, each of the C02 bombs may be detonated to release the C02 within and engulf the fire in C02 and steam from the ice.

[00206] Figs. 15 and 16 show an example of a driverless vehicle as a front view cross section and a side cross section respectively.

[00207] The driverless fire control vehicle 1501 may be used in the emergency response system as described herein. The driverless fire control vehicle has a body with a heat reflective exterior 1503. The exterior may be a shiny metal exterior to reflect radiated heat. A water blanket may be provided to cool the outside wall of the body. For example, circulated water may be provided to keep the water blanket cool. A steam release valve 1517 may also be provided to release steam that has been generated as the water blanket is heated. The driverless fire control vehicle may also have heat reflective and heat insulating panels that can be positioned to fill the space between the ground and the body of the driverless control vehicle to protect the tyres, wheels, engines and/or the underside of the driverless fire control vehicle.

A system that will use water inside the vehicle to cool the exterior of the vehicle can also be added to the driverless fire control vehicle, with an example being the spraying of water onto the outside of the fire control vehicle.

[00208] The vehicle may have one or more engines or motors 1505. Multiple wheels 1507 driven by the engine are provided. The wheels may have replaceable tyres in the event they are damaged. The wheels are steerable. [00209] The vehicle has a control module 1509 for controlling the different components in the vehicle based on control signals received via a communication module 1511 that communicates with the emergency response system using any suitable communication protocol.

[00210] A fire-retardant release system 1513 is provided for releasing a fire-retardant material. For example, the fire-retardant release system may be a fire hose nozzle that is connected to a hose 1515 that is rotatable via a motor (or the engine) to turn 360 degrees.

[00211] The communication module is arranged to receive control signals, transmitted for example via the server of the emergency response system, to enable the control module to control the engine and the fire-retardant release system.

[00212] The driverless fire control vehicle may have one or more fire control vehicle sensors (1601), which may include one or more of a heat sensor, humidity sensor, moisture sensor, smoke sensor, wind direction sensor, wind speed sensor, lightning detection sensor, a visual sensor, an audio sensor, a location sensor, an orientation sensor, and a magnetic field sensor.

[00213] The communication modules in the land based emergency response vehicles may be shielded, for example, the satellite receiver may be covered with a cover that is capable of withstanding high temperatures and the land based emergency response vehicles may have a retractable Wi-Fi and mobile phone antennae.

[00214] Various alternatives, additions, example algorithms and scenarios for the emergency response system as a whole are now described.

[00215] Other types of sensors may also be used for different types of emergency situations to which the herein described emergency response system may be applied. For example, the sensors may include a person detection sensor, an animal sensor, a water level sensor, a snow sensor, a chemical sensor.

[00216] These types of emergency situations may include, but are not limited to, a potential drowning, flooding, avalanches, earthquakes, dam leaks, chemical leaks tec.

[00217] For example, the location of the emergency situation can be determined from one or more sensors, such as for example, a person detection sensor, an animal sensor, a water level sensor, a snow sensor, a chemical sensor. That is, the emergency variables provided by these sensors can assist in the server determining the location of the start or continuing development of an emergency situation.

[00218] As discussed before, the server may determine what kind of emergency situation (i.e. what type of emergency situation) is occurring can be determined by the server by analysing the emergency variables received from the sensors and comparing those variables with known scenarios to see if the received variables match the variables for known scenarios.

[00219] For example, the kind of emergency situation may be a chemical leak, where the emergency variable is received from one or more sensors, such as for example, a chemical sensor and a location sensor etc., where, if the emergency variable (detection of a particular chemical signature) matches or is within a defined threshold, the server may determine that a chemical has been released.

[00220] For example, the kind of emergency situation may be a person lost at sea or in the bush, where the emergency variable is received from one or more sensors, such as for example, a person detection sensor and a location sensor etc., where, if the emergency variable (detection of a person in a potentially distressed state) matches or is within a defined threshold, the server may determine that a person is in distress and needs assistance.

[00221] For example, the kind of emergency situation may be a flood, where the emergency variable is received from one or more sensors, such as for example, a water level sensor, a visual sensor, an audio sensor and a location sensor etc. where, if the emergency variable (a water level in metres for example) matches or is within a defined threshold, the server may determine that a flood is occurring. This information may be transmitted to the server and then made available to the public via the server so that they can plan travel on open roads and effective evacuation if required. This information may also be provided to online mapping services to make travel planning more efficient.

[00222] It will be understood that the analysis of the emergency variables may be carried out at a computing device other than, or in association with, the server.

[00223] It will be understood that the output of the determination to the resources may be communicated directly from the server or via one or more other communication devices such as those positioned on or forming part of, for example, the drones, the embedded electronic device, the sensing unit, the sensing tower, the driverless fire control vehicle, or any other suitable part of the emergency response system. [00224] Various databases may be utilised as described herein to store data for use by the system in making determinations.

[00225] According to one example scenario, the system may utilise one or more sensors to detect a fire. Upon the system determining that the fire requires immediate attention and upon determining that the severity is high, the system may output instructions to release a swarm of drones (as described herein) that can be used to monitor and/or attack the fire. This enables the system to direct a large-scale attack of the fire at very short notice. As described herein, the drones may communicate with each other as well as the system in order to co-ordinate the attack and communicate with surrounding resources and/or the public. Further, as an alternative, or in addition, a large number of automated vehicles (as described herein) and/or manual vehicles may be utilised to monitor and/or attack the fire. Manned aerial vehicles may also be used as an addition resource provided that the drones are kept at a safe distance from the manned aerial vehicles.

[00226] Various other scenarios are described, or previously described scenarios are elaborated on as follows.

[00227] The information and data (e.g. emergency variables) used by the system can be collected from sensors in fixed locations, such as on cliffs, mountain tops, other natural structures with a good view of the surrounding country, towers, including mobile phone towers, transmission lines, water supply infrastructure, buildings etc. Also, the information and data can be collected from moving locations such as aeroplanes in flight, drones, land vehicles, people with specifically designed mobile phone Apps, satellites, low earth orbit satellites (some of which cannot see through clouds in the visible spectrum), databases and services managed by organizations like the Bureau of Meteorology (BOM), GIS systems which hold accurate information about the terrain, the infrastructure, location and importance of structures. Also, information on whether the land is farmland and likely to have livestock is provided as well as information on road and rail access, location of airports, airstrips and drone recharging stations and so on. The system is expandable so that it is fast and easy to add additional information and sensors when needed.

[00228] The system examines the infrastructure to see what needs to be improved and made more secure. With higher temperatures and strong winds, transmission towers are collapsing. Reinforcing the towers with additional steel and possibly with guy wires (without causing dangerous induced currents in the guy ropes) may reduce damage to the towers and the transmission lines. New construction may be aimed at survivability, e.g. by constructing new towers to a much higher standard and/or burying the cables in the ground.

[00229] Information from inaccessible places may also be very useful. Helicopters can lower self-contained, solar powered sensors (e.g. sensing towers) onto inaccessible cliff tops, and into the ground in forests so that soil moisture and fuel load can be measured. That is, the lifting point 512 on the sensing towers enables the towers to be transported. The sensors may be powered by batteries and solar. In some cases, the base of a tower may be constructed using humans lowered into the location and the tower lowered onto the base plate. One possible installation is to have a cone on the base plate onto which the sensing tower is lowered. It would be possible to move these detectors from one location in winter, when fire is low probability, to an area where flooding is possible, and then return them in time to that location for the next fire season.

[00230] Another feature of these sensor systems is that they can provide remote automated docking and charging for drones which would require a more substantial solar installation. Drones would not be flying except when locating lightning strikes and fires or collecting information for emergency response classification and response. Information provided by the BOM database may cause the server to enable the drones to be put on standby. The drones may also be controlled to fly when there is a need for the drones to collect other information.

[00231] In addition to, or instead of, solar charging, batteries can be provided to power the sensors and also to power drones. These batteries can be manually replaced by ground or helicopter access.

[00232] The emergency response system can also assist in preparation for emergencies e.g. by calculating the costs of the loss e.g. of a transmission line to the community, compared to the cost of monitoring the transmission line.

[00233] This analysis can be extended to understanding the value of land and structures that should be protected ahead of other areas that are uninhabited. This information may assist making more informed decisions when deploying scarce resources.

[00234] The information that can be collected can assist in a large number of emergency situations, such as bushfires, by locating fires very early so that these fires can be extinguished or suppressed quickly, measuring floods and water inundation, seeing which roads are open and available to use for supplies and evacuation etc, assessment of damage after storms and prioritization of the response, finding people lost in the bush or at sea during the day and at night, delivery of urgent medical supplies like insulin or antivenom, and so on.

[00235] Surveillance can relate to other emergency situations other than fires. For example, the system may look for and locate people at sea or on land using visual or audible sensors, or look for and locate people who get lost in the bush or who are abducted etc. by alerting people on the App. The system can check the state of transmission lines and roads in good weather using the sensors. The system can monitor agricultural land. The system can be used in natural disasters to get immediate and accurate information such as heights of rivers etc.

[00236] The data (e.g. emergency variables and instructions) can be transmitted by using one or more communication protocols or methods, such as the mobile phone network, telecommunications cables (phone lines, coaxial cables, optic fibre cables), satellites, especially the proposed low altitude satellites, power lines, line of sight communication technologies such as FM radio and microwave technologies, and so on. A drone, or a swarm of drones, flying high enough to communicate with intact phone towers can be used to provide communication resources when local communications have been brought down by the emergency. In addition, planes, drones and vehicles can use satellites such as that proposed in the SpaceX Starlink project.

[00237] An App for a mobile phone can be downloaded and enable the phone to be registered to a person using the phone. The registered users can take photos, classify these photos by drop down menus or by written description, and the App will automatically record the latitude, longitude, elevation, direction of the phone when the photo is taken, angle of tilt of the camera, temperature and any other information that is available in the hardware of the device running the App.

[00238] Many bushfires are accidental: people starting barbeques, welding or angle grinding outdoors and so on. These fires can get out of control quickly. It is important that the fires are detected quickly to assist in fighting them. Accidental fires usually start around human habitation. Providing people with an App that can report a fire and deliver precise information to the system to enable a response by resources, such as the Fire Brigade, as it is close to human habitation is essential. If the Fire Brigade is not available or if there isn’t a Fire Brigade locally, as determined by the system, then the system may determine that an aerial response may likely be needed and the system can prioritize and schedule this response. [00239] The App can also be used for people to say that they are going on a hike and show the route, the departure time, the arrival time, and every half hour the App would send back the latest position and the state of the battery of the mobile phone to the server. The system would then send a message to the hiker to check off that they are safe when the hiker arrives at the destination.

[00240] If there is little recent data from along the route, the system may send a message requesting that the hiker takes photos at specified locations where other photos have been taken so that the changes can be analysed. This may provide useful information about fuel loads and vegetation.

[00241] People going to sea or flying can also provide similar travel plans and have these communicated to the server.

[00242] If the travellers have not arrived at the destination at the appropriate time (with a defined threshold), drones can be dispatched by the system to the last reported location and can take suitable equipment and supplies, such as floats with waterproof GPS locators for sea travel, water, food, protective gear and a communications device and external phone batteries for hikers, and suitable equipment for potential air crashes.

[00243] Microphones and loud speakers on drones may also assist to find people, such as lost hikers. The drones can also provide communication to people who are out of cell phone range by providing long range WIFI linked to satellites, and the loud speakers can be used to transmit messages to them to let them know that there is a Wi-Fi communication option and to switch on their phones. Alternatively, the drone may send a message to the App in the phone to start communications with base.

[00244] In addition to satellite communications, drone to drone communications using long range WIFI is envisaged.

[00245] The system may also contain manuals that detail the response to each different type of emergency. Given that over 40% of the Australian population read at grade 5 level or less, this information must be prepared and displayed to maximize the ease of reading. Diagrams set out according to the principles of cognitive load theory should be provided as well as audio and videos. [00246] Various different sensors will be used to simultaneously collect and analyse relevant information for the emergency response system. This will include manned or unmanned observation (sensing) towers, satellites, high altitude drones, planes, helicopters and drones, humans with Apps, sensors in the terrain such as forests and in the grasslands, and so on.

[00247] Sensors to detect meteorological information such as humidity, temperature, wind direction, air pressure etc can be used to gather useful information, such as within a forest, that can be used to predict the likelihood of fire and its rate and direction and speed of propagation. Sensors in the soil can detect moisture content.

[00248] The bigger the fire, the easier it is to see. A small fire can be easily extinguished if it is attacked in minutes but after 30 minutes, it may be out of control. An out of control fire will rapidly get into the canopy of a forest and will be very easy to see from the air or from satellites. The described system provides a way to detect small fires quickly and accurately. The accuracy is needed so that planes and/or drones are not diverted from other real fires by false alarms.

[00249] Some lightning strikes may light tree canopies, and these will usually be visible from the air. Different canopies may emit different light spectra and this could provide useful information about the heat of the fire and this, together with other variables like weather conditions, humidity, terrain etc along with historic data on the rate of spread of fires may enable estimates of the spread of the fire to be calculated.

[00250] Grass fires can be readily detected from the air as there is nothing hiding the fire. Tree canopies can obscure the early detection of fires. The denser the tree canopy, the more the tree canopy can obscure a small fire. Mapping the tree canopy density will provide the system with information that can be used to optimize flights over the area. For example, if the canopy is permeable to fire radiation, then using above canopy drones travelling quickly may be the best way to survey that areas of the forest. However, if part of the forest has a think canopy, then sending under canopy drones will provide a better solution. Drones travelling in a forest may need to travel more slowly than above the canopy, and so may take longer to arrive at the scene of the potential fire.

[00251] Experimental data from experiments with sensors may be used to develop the best way to sense fires through canopies. For example, it is possible that different trees in the canopy will let through different radiation compared to other species of trees. Mapping the species of trees together with the density of the canopy and matching this information with the experimental detection capacity of the sensors may provide the system with the capacity to plan an efficient route for the fastest available aerial resource to locate and observe the fire.

[00252] Drones that can see through a canopy will be able to detect fires underneath the canopy. Experimental data from experiments may be obtained by lowering instruments beneath the canopy that will exhibit the characteristics of a fire, such as smoke, and infrared and visible light without the risk of a fire. Instruments on the drone and on other nearby drones may measure the smoke and infrared and visible light to measure the permeability of the forest canopy for the detection of fires from above the canopy. Smoke and low temperature LED visible light released by the instruments will be unlikely to start a fire. Infrared light will only be turned on where there is sufficient distance from inflammable materials to be safe.

Experimental data may be obtained from experiments conducted at times when the weather will make the starting of fires difficult, e.g. low temperatures, high air humidity, damp fuel etc.

[00253] Photographs of the canopy can be matched with experimental measurements at those places where the photographs were taken to enable machine learning algorithms to be developed to calculate the probability that a fire is detected correctly for action, no fire is detected correctly for no action, no fire is incorrectly labelled as a fire for action, and a fire is incorrectly labelled as no fire for no action.

[00254] A second impediment to accurate observation is the existence of undergrowth.

[00255] One strategy is to locate a drone down wind of the possible fire to let the wind blow the smoke to the drone. Experimental data may be obtained from experiments that are developed to see if the undergrowth in that region will create smoke. For example, in very dry and hot conditions, some materials will burn with very little smoke, so the probability of an accurate fire detection using smoke may be compromised. This information may be tested with various sensors to allow a suitable sensor to be selected that will work accurately in the field. Knowing the build up of undergrowth in the forest and the weather conditions may enable the system to determine whether looking for a fire with multiple drones is a good allocation of resources.

[00256] The state of the undergrowth and the permeability of the canopy may mean that a drone is unlikely to detect a fire for some time. Experimental data from experiments with drones detecting fires in controlled environments may be used to measure the elapsed time between a lightning strike, for example, and the size of the fire that the drones can easily and accurately detect. This timeframe may be used by the system to calculate the most efficient way to schedule drones, so that drones are not sent to inspect before they will detect the fire and the elapsed time is not so great that the fire gets out of control.

[00257] Experimental data from experiments may be used by the system to determine the best time for inspection of different terrain and undergrowth to enable better scheduling of inspections to enable more accurate observations.

[00258] Light and infrared radiation can be analysed e.g. by a Fast Fourier Transform that will show the frequency spectrum of the radiation which can be matched against known frequency spectrums to better understand such things as what material is burning, the type of fire (undergrowth, canopy, undergrowth and canopy etc) and the temperature of the fire.

[00259] Drones that can fly underneath the canopy may be used to take 360 degree videos of the ground to provide the system with detailed information about the build up of potential bushfire fuel on the forest floor at specific locations. These photographs or videos can be compared using machine learning systems to photographs or videos where the energy density is known to enable an estimate of the fuel loading to be calculated.

[00260] Th system may determine that under canopy drones may be required when the canopy can stop the detection of a fire in the lower canopy or in the undergrowth.

[00261] The smaller the inspection drone, the easier it will be to find holes in the canopy. The lightning strike itself may open the canopy and this can be detected by the drone when it finds the lightning strike from above by using machine learning systems that have analysed places where lightning has struck and punched a hole in the canopy.

[00262] Digital maps may be created and used by the system to show clearings in the canopy that will allow a drone to drop beneath the canopy through the clearing. The information for the clearings can come from a variety of sources such as satellites (low orbit and geostationary), solar powered gliding drones that can stay aloft for months, planes, drones and from the ground.

[00263] Drones can be used by the system to map routes underneath the canopies so that the under canopy path to a potential lightning strike can be chosen to minimize the time to the potential lightning strike. [00264] Control systems for the drones can use 3D visual images of the forest for navigation in daytime and using strong lights at night time. The problem in a forest is simpler than navigating through traffic in streets. The tree trunks do not move although the canopy does. However, branches in the canopy can and do move, but the drones will not fly through the canopy except to ascend and descent through it. The visual system is looking for empty space in the direction of the fire, and avoids obstacles, such as trees. The speed that the drone can move will be in part determined by the number of trees in an area and by the size of the drone. Small drones should be able to find an opening in the canopy close to the suspected fire and then rapidly move to the fire location.

[00265] It is extremely important to detect fires as soon as possible. A fire can probably be put out after a few minutes with a small quantity of water or a fire control device. The difficulty in extinguishing the fire increases with each additional minute.

[00266] In bushfire areas, sensors can be used to detect heat from fires, visual identification of flames and smoke, noise of a fire (e.g. crackling), identification of the fire from the colours of the flames (think spectral lines), motion of the flames, and the use of other detectors such as identification of smoke using an ionization detector which employs a radioactive material to ionize the air in a sensing chamber. The presence of smoke affects the flow of the ions between a pair of electrodes, which triggers the alarm. They can detect fires with little smoke. These detectors could be placed in areas where they are likely to encounter smoke due to prevailing winds, topography of the terrain etc as a backup if a fire was not detected if caused by lightning, or was caused by other means.

[00267] Many bushfires are caused by lightning. Lightning is hard to predict but some new tools may allow lightning to be predicted in a 10-30 minute timeframe in 30km area.

[00268] Lightning arises because a positive charge is generated at the top of the cloud and a corresponding negative charge is generated at the bottom of the cloud. This gives rise to four kinds of lightning: Between clouds (CC); Intracloud (IC); Lightning balls but these are very, very rarely observed and so are not prioritized; Clouds to ground (CG).

[00269] There are two kinds of CG, negative lightning is where the electrons from the base of the thunderstorm descent to the earth and positive lightning where the electrons from the earth ascent to the positively charged cloud. [00270] Sensors may be used to distinguish between positive and negative lightning strikes. Positive lightning is usually composed of one stroke, while negative lightning often has two or more strokes. This requires a fast camera to detect.

[00271] Monitoring for positive lightning strikes is far more complex as they may occur over 30 kilometres from the parent thunderstorm. This will require multiple monitoring resources from towers to satellites to drones.

[00272] Location of the strike, vegetation at the strike site, weather conditions at the site, proximity of the strike to the parent thunderstorm, whether there are two or one flashes, and the intensity of ELF and VLF; all of this information may be used as emergency variables by the system to calculate the probability of lightning being positive or negative lightning, and the probability of the lightning strike causing a fire.

[00273] If there are sufficient resources, then the inspection of all lightning strikes may be undertaken both to record data about what the strike looks like and to ensure that there is no fire there. It is only if there are more lightning strikes than resources to inspect and attack potential fires will there need to be active prioritization of the different sites by the system.

[00274] The system analyses and detects where lightning strikes occurred as accurately as possible to reduce search areas, and as close as possible to real time. Two installations of cameras may be used in sensing towers that can detect the angle of a flash can measure where the lightning source intersected the ground. Three installations of cameras may also be used to be more accurate.

[00275] Camera installations that have a significant distance between cameras may be used for increased accuracy than cameras that are close to each other, as the distance between the cameras will improve the 3-dimensional accuracy of the image.

[00276] The lightning detectors may be fixed in high risk areas, or can be moveable, on wheeled vehicles moved into position following a thunderstorm, or on manned planes or on unmanned drones. Metal planes can fly in thunderstorms as the metal in the plane forms a faraday cage. These planes can be fitted with cameras and GPS location to allow the location of the lightning intersection with the ground to be accurately calculated. The drones are made lightning proof and any equipment inside is properly shielded in faraday cages, such as using conductive plastic for example. [00277] Winged drones may have cameras mounted on the wings and can be separated by substantial distances. These cameras may have different fields of view. Forward facing cameras may have a narrower field of view so that they can detect forward facing events with greater accuracy. Wider angle cameras can be mounted on the rear of the wings and on the fuselage facing parallel to the wings, so that any flashes are detected. Very high-resolution cameras may be used to enable the systems to zoom in on the image to detect the flash and calculate the location of the flash.

[00278] Artificial intelligence (machine learning) may be used to improve the accuracy of the location of the lightning strike and reduce both the incidence of false positives and negatives. For example, if the system detects lightning from 3 different observation (sensing) towers striking the ground, the system may triangulate the place where the lightning “hit”. The system may then send a drone to the detected location to inspect the area using visual sensors (or other sensors) and find the actual location. This information may be provided to triangulating software via Al to better improve the prediction of the place where lightning strikes. Other variables like terrain and geology may also be included.

[00279] Filming from the air locations where lightning has struck will enable the development of lightning site recognition software based on machine learning of sites where lightning has struck and comparing to sites nearby or in other areas where lightning has not struck.

[00280] According to one example, a lightning detecting sensing tower may have a 360 degree view of the valleys using an octagon layout of 16 detecting cameras with 2 cameras at the edge of each side of the octagon. A second octagon may be situated above the first octagon but offset 22.5 degrees so that the vertices of the lower octagon are in the middle of the sides of the higher octagon. The direction of the lightning flash can to some extent be determined by analysing the images to determine which camera detected something and where. The top octagon may be separated from the bottom octagon by 2 metres or more to provide more accurate 3D resolution of the lightning image. Several of these sensing towers in line of sight may provide a more accurate estimation of the location of where the lightning struck the ground. The more accurate the location of the lightning strike, the faster it becomes for the system to utilise resources to examine the location for evidence of fire. For example, one drone may be used to examine more lightning strike locations in a given timeframe.

[00281] Once the location of the lightning strike has been determined by the system, the system then determines whether the lightning may have caused a fire using the available sensors. A fire is more probable with dry lightning, i.e. lightning without rain. There are at least two possibilities: take another observation to see if a fire has started, or utilise resources, such as a water bomber or a different fire suppression device that can be delivered by the air, to the area where the lightning strike occurred without an additional observation.

[00282] The decision probabilities can be calculated by the system based on one or more of the following emergency variable inputs:

[00283] What are the fire risk conditions as determined by the fire authorities? Catastrophic, extreme, severe, moderate etc.

[00284] What is the likelihood of significant rain in the location in the very near future?

[00285] What are conditions like in that location, such as the dryness of the fuel load at that location?

[00286] What is the likelihood that a fire will start from the lightning strike in that location in the current conditions, calculated from historic data? For example, if there are 4 lightning strikes in an area and only one fire starts, then the simplistic probability that in similar weather conditions, a fire will start will be 25%. The addition of different parameters, such as the terrain, the vegetation at the terrain and the lightning type may adjust the probability calculation based on existing data.

[00287] What is the risk of a false positive lightning strike reading at that location (no fire but reported as a fire)? Near a town, flashes of light could come from human activity, but this is less likely deep in an inaccessible forest. The probabilities of false positive can be reduced by using machine learning software. Drones may be sent out by the system behind thunderstorms to detect where lightning strikes and then sense whether the strike started a fire.

[00288] What is the risk of a false negative (a fire that is not reported)? Does this vary for different terrain - e.g. a forest canopy may make fires harder to detect? The cost of a false negative may be catastrophic. Accuracy may be improved by calculating optimal inspection times, using multiple sensors, calculating the probability of a fire caused by lightning in other similar areas and so on.

[00289] Can the observation be made quickly, e.g. by a manned plane or an unmanned drone in close proximity to the lightning hit location, without diverting the plane from following the thunderstorm? Fast drones may be able to collect and analyse data in considerably less time than slower drones, and mean that fewer drones are needed to provide surveillance for an area.

[00290] The system can determine how accurate its assessment is of whether a fire has started or not. An assessment is made by the system as to whether there is a fire. If the assessment is that there is no fire, and if there is a fire, the fire will almost certainly burn and get bigger if the conditions are right. The bigger the fire, the easier it is to see. So, provided that there is subsequent surveillance in a defined time period, such as 1 hour for example, of the where there was no fire, the system can determine whether its assessment is correct as there will either be a fire visible from above the canopy, in which case the determination was wrong, or there will be no detectable fire, in which case the determination has a high probability of being correct. The surveillance information will be used to improve the prediction accuracy of the system.

[00291] The system may determine the opportunity cost to following the thunderstorm or diverting the lightning detecting plane or drone to see if a fire has started, or instructing the plane or drone to inspect the location of the lightning strike without interfering with its ability to detect the next lightning strike.

[00292] The system can determine whether there is a high demand for water bombers at that moment. If there are water bombers available then bombing the location of the lightning strike within a few minutes is likely to be the safest option to use by the system.

[00293] The system can determine whether the water bomber is fitted with a fire detector so that the water bomber can inspect the lightning strike location and determine if there is a fire before bombing it.

[00294] The system can determine whether there is a water bomber in the vicinity that is available. If there is, and if say the lightning strike were observed from say fixed cliff based sensing towers, then sending instructions using the system for the water bombing at a lightning strike location straight away may be the least risky action if that bomber could not rule out that there was a fire started.

[00295] Various strategies may be used by the system to avoid false positive readings, including: [00296] In addition to light, the flow of electrons in lightning causes the air to heat up to approx. 30.000C, which produces additional information that can be sensed. One of these sensors is infrared detector. When a radio frequency (RF) lightning signal is detected at a single location, one can determine its direction using a crossed-loop magnetic direction finder but it is difficult to determine its distance. The crossed-loop magnetic direction finder may be useful if there are two or more fixed locations allowing the position to be triangulated.

[00297] A lightning discharge generates both a RF electromagnetic signal - commonly experienced as "static" on an AM radio - and very short duration light pulses, comprising the visible "flash". A lightning detector that works by sensing just one of these signals may misinterpret signals coming from sources other than lightning, giving a false alarm. Specifically, RF-based detectors may misinterpret RF noise, also known as RF Interference or RFI. Such signals are generated by many common environmental sources, such as auto ignitions, fluorescent lights, TV sets, light switches, electric motors, and high voltage wires.

[00298] Likewise, light-flash-based detectors may misinterpret flickering light generated in the environment, such as reflections from windows, sunlight through tree leaves, passing cars, TV sets, and fluorescent lights.

[00299] However, since RF signals and light pulses rarely occur simultaneously except when produced by lightning, RF sensors and light pulse sensors can usefully be connected in a “coincidence circuit” which requires both kinds of signals simultaneously in order to produce an output.

[00300] The lowest cost strategy may be to use visual cameras separated by say 500mm to enable the cameras to estimate the location by each having a direction of the flash which will have slightly different angles and the distance to the object can be calculated (parallax rangefinder and stereopsis - using binocular vision.) Having multiple cameras on one sensing tower with different horizontal and vertical locations can improve the accuracy of the distance estimation.

[00301] Multiple sensing towers may also provide additional information and accuracy be enabling triangulation.

[00302] Lightning proof drones will have cameras mounted on them with 360 degree view of the world, optionally at different heights, that enable the system to accurately estimate the location of a lightning strike. For example, if a swarm of drones are following the thunderstorm, they will be able to triangulate the strike as well as find the strike location straightaway.

[00303] A drone may be sent to inspect the areas hit by lightning and to find the exact location of the lightning strike, where the location may be added to the lightning strike database to improve the accuracy of the system.

[00304] The system may utilise a large number of detecting devices to provide system redundancy, and allow for all possible fires to be observed and reduce the need for larger bombers to be deployed on unnecessary fire suppressing attacks.

[00305] If there are more detecting devices than there are locations which need to be inspected, scheduling is easy as detecting devices may be sent to any potential fire site. In catastrophic fire conditions with multiple dry lightning strikes across thousands of kilometres, resources are likely to be stretched.

[00306] Therefore, existing planes and helicopters may be fitted with detectors, and drones or drone swarms can be used by the system to inspect potential fire sites be developed.

[00307] A combination of three different types of sensor (detector) may be used in detection of fire, for example. Light, infrared and smoke detectors may be used. A positive identification of fire from all three provides a very high probability there is a fire for the system to make the determination. Smoke detectors may be positioned down wind and they may not detect anything until a sufficient amount of smoke has been made and dispersed by the wind. Therefore, positive detections from light and infrared sensors from an upwind position will mean there is a very high probability of a fire, resulting in a positive determination by the system. Similar results may also apply for false negatives.

[00308] Installing computing resources on aerial resources may enable the compression of signals communicated to and from the system, and allow the analysis of information collected by the sensors. For example, fast Fourier transforms may be performed on visual and infrared heat images obtained by the visual sensors. Different vegetation may produce different Fourier transform patterns when burning at different temperatures. For example, in many instances, the canopy burning will produce different Fourier transform patterns than the undergrowth.

Matching of the transformed information to stored transform information collected experimentally from fires may enable the system to estimate the vegetation burning, the temperature of the fire and may provide critical information about how to respond to the fire, as responding to a canopy fire may need different resources to those needed when responding to an undergrowth fire.

[00309] Fires may be caused by the transmission of electricity. For example, a spark sensor (detector) may be used to sense a rapid change in infrared light coming from a tiny hot particle traveling at high speed through its field of view. Sparks may be detected by mounting spark detectors on transmission lines to enable the system to detect sparks quickly. Alternatively, other sensors, such as cameras and smoke sensors may be used.

[00310] The system may have an increased number of aerial response resources to enable a better response to emergency situations, such as all fire situations and especially to catastrophic fire situations.

[00311] Larger drones may be used to deliver fire retardants (e.g. C02 bombs) and other devices to attack fires.

[00312] The drones may utilise a vertical take-off mechanism so that they can have a base close to the emergency situation.

[00313] Drones may be powered by batteries, petrol or any suitable power source. The drones may have battery packs that are replaceable in order to quickly recharge the drone. The battery packs may be located close to the emergency situation. Refilling drones with petrol may be done anywhere.

[00314] Large aerial tankers may be used as a resource to drop smaller amounts of fire retardant on a number of smaller fires.

[00315] Crop dusters may be upgraded and included as a resource so that they can also function as water bombers.

[00316] Small planes may be a resource where they are adapted to drop water - e.g. by removing seats and putting in a large water container in the luggage hold and installing a water release mechanism under the body of the plane.

[00317] Planes may be a resource where they are adapted to attack fires with fire retardant (e.g. C02 bombs) and other devices. These planes may include additional instrumentation. [00318] The system may provide improved accuracy of detecting the emergency location (e.g. fire location) and the effectiveness of the response by a single emergency response, e.g. so that more and more fires are extinguished in one response.

[00319] Pilots may be classified so the system knows whether they can fly at night, in poor visibility, with instruments etc.

[00320] The system may require active redeployment of aerial resources to areas of great fire risk, determined in part by rainfall to date, current weather conditions and weather predictions.

[00321] The system may require that planes and helicopters are already in the air on catastrophic fire days so that they can bomb a fire in a very, very short time. On less extreme days, they can be positioned on strategically located airports or air strips and be ready to scramble.

[00322] If fires can be caught early, then the plane or helicopter can be instructed by the system to drop the water with flame retardant directly over the fire at low altitude with little risk. The bigger and hotter the fire, the more risk there is.

[00323] The system may understand that rugged remote areas may be better defended by helicopters than planes, as helicopters may be able to get closer to the seat of the fire for a more targeted hit.

[00324] The system may include a large number of strategically placed centres where helicopters and planes can refuel and take on water with flame retardants so that there is a short travel time between sorties. These may be landing strips for smaller planes. Helicopters can access water in dams but it may be more efficient to have water and flame retardant stored in strategic locations so that the number of sorties can be maximized in extreme weather. In bigger areas, the system may include local refuelling.

[00325] The system may have fast reloading facilities for helicopters: the system may have two above ground structures and/or two holes in the ground that have filled water containers with flame retardants. The containers may be a suitable fabric held up with a number of ropes attached to the edges. A helicopter may be instructed by the system to drop the used fabric container and hover over a structure or a hole where a filled container is attached to the helicopter. The used container is put in the hole and filled up for the next trip. Having multiple fire retardant containers lifted by a helicopter may mean that multiple small fires can be responded to on the one sortie.

[00326] The detectors and other equipment in the manned planes and helicopters are automated where possible so that the pilots are able to concentrate on flying the planes and avoiding danger in a perilous environment.

[00327] Whenever a plane, helicopter or drone is instructed by the system to apply fire suppressant measures the resource may film the area involved from above so that this visual data (emergency variable) can be used to improve accuracy of lightning strike location and subsequent fire detection, for example.

[00328] Where the system utilises smaller planes and drones to quickly deliver effective fire suppression, light weight effective fire suppression devices may be used that can be very accurately dropped from a plane, helicopter or drone to suppress or extinguish a fire. Fire suppression will slow the expansion of the fire and enable more time for larger water bombers to be instructed by the system to arrive and water bomb the fire.

[00329] A non-flammable device that is light weight may be provided, that can be dropped from a manned plane or helicopter. The device may be delivered automatically by a low cost, unmanned electric drone that can also inspect the site of a lightning strike to make sure that there is a fire that needs to be suppressed or extinguished. Machine learning technologies may be employed to enable the drone to locate the best place to drop the fire suppression device and the best direction to approach the drop, so that the fire suppression device can quickly be dropped on, or flown or glided into a fire to extinguish the fire or significantly slow its spreading a light fire suppressor that can be quickly and accurately delivered by a low cost drone. A version of the light, drone delivered, fire suppressor may be released remotely and fly or glide into the base of the fire. These devices can also be transmitting useful data about the fire back to the drone, which in turn can then transmit this information to the server of the emergency response system.

[00330] The server may utilise data associated with how best to respond to the emergency response, e.g. a fire and where to attack the fire, based on historic information and data produced from controlled experiments. Multiple controlled experiments may be performed and data collected in different circumstances to train the system. The system will look up the closest match to the existing circumstances and use the associated data to determine how to respond to the emergency response and which resources to use. During and after the emergency response, the response and the effect of the response may be recorded, and this data may be used to train the Al systems with how to optimize the response in similar circumstances.

[00331] The system determining the most vulnerable place to attack a fire coupled with accurate aiming devices enables a light drone delivered fire suppressor to be highly effective in suppressing or extinguishing the fire. Different suppressor designs may be used for different fires. The system may use data from a series of experiments conducted to determine the most effective way to extinguish or suppress a fire, based on the terrain, the undergrowth, the forest cover, the weather conditions and the direction and place of attack. This information may be used by the system to direct the response in similar circumstances. During and after the fire response, the effect of the response will be recorded, and this information will be used by the system to better train the Al systems with how to optimize the response in similar circumstances.

[00332] An alternative approach is to have a single, larger suppression device on a larger drone that will extinguish many fires and suppress bigger fires to enable more drone to attack the fire or call in a larger water bomber.

[00333] The system may utilise one or more sensors to determine whether there are any humans near the emergency response area. The system may map areas that are inaccessible and are unlikely to have humans underneath. Th system may use cameras to look for humans. The system may use special sensors to look for humans e.g. using infrared. The system may use loudspeakers and microphones to check there are no humans. The system may use suppressors to explode above the fires and not have fragments that can hurt people. Even if there are people there, they may be threatened by the fire and it may be safer to attack the fire with e.g. a C02 bomb than not attack the fire or delay the attack until a water bomber is available.

[00334] A helicopter may be instructed by the system to fly directly above a small fire at low altitude, reduce speed to nearly a hover and then drop water directly onto the fire. If there is wind, then the helicopter may be instructed by the system to hover in a position so that the wind carries the water onto the fire. For this to happen, the system may need to be notified very quickly that a fire has started, know exactly where the fire is, and direct the helicopter to arrive there quickly. [00335] If the fire is bigger, then the helicopter will be instructed by the system not to hover over the fire for safety reasons. The resource nay be instructed to attack the fire and drop the water at a higher altitude up wind so that the wind carries the water onto the fire.

[00336] The response may reduce the heat of the fire, and this may enable the helicopter to attack the fire directly above the fire. If the helicopter has two reservoirs of water, it could be instructed by the system to drop the second reservoir directly over the fire (assuming no wind) at lower altitude.

[00337] The system may use research data based on the best way to attack fires in different terrains with different fuel load and canopies. This data may be used by the system to calculate the best approach for the helicopter and may be transmitted to the pilot by the system to maximize the efficiency of the attack.

[00338] A similar approach may be used for larger tankers. These tankers may be instructed by the system to drop only a small amount of load directly onto a small fire. This will allow a large tanker to potentially put out multiple smaller fires in one response.

[00339] The planes may be equipped with cameras and other sensors so that the effectiveness of the attack can be estimated from captured data. Drones can also be used to provide data on the effectiveness of the response. A library of images of successful attacks will be stored by the system to match successful attacks with the attack response, e.g. direction relative to the wind, height, accuracy of the drop, number of drops etc.

[00340] The App on the mobile phone under control of the server may inform a pilot about efficient fire approach routes and exactly when to release the fire suppressant device can be constructed by using sensors built into the mobile phone, and by connecting with sensors on the plane. Mobile phone sensors may include time, elevation, position, travel direction, temperature etc. Information from sensors on the plane communicated via Bluetooth may include altitude, wind direction and speed.

[00341] The App may calculate the time when the pilot should release the fire retardant bomb based on location, velocity, height, time to drop etc. For example, assuming the plane is travelling at a velocity of vp and at a height df above the ground where the fire is. Assume that the bomb is to detonate at a height dd above the fire. The distance that the bomb will fall is df- dd. [00342] The time to fall from df to dd (df-dd) can be calculated. Let’s call t=(2(df-dd)/g)**1/2 if there is no air resistance. As the bomb starts from zero speed vertically and accelerates slowly, this is a reasonable approximation. The bomb will likely have a horizontal velocity as it will be dropped from a plane or drone. The distance dp travelled by the bomb due to the plan’s velocity will be dp= t x vp. If there is a wind, then the distance that the bomb will move along the direction of the wind dw which can be calculated as dw = t x vw where VW is the velocity of the wind. The actual distance moved by the bomb may be calculated by resolving the wind and plane speed vectors, taking into account the direction of the vectors and the angle between the vectors. Experimental data collected from experiments dropping fire retardant bombs may be used by the system to determine the relevance of drag to the bomb aiming, and to enable the system to determine whether the bomb aiming devices need to compensate for the drag.

[00343] The plane may be instructed to drop the bomb at a location that will deliver the bomb onto the fire that will be calculated by resolving the vectors dw and dp and the resulting effect of the bomb velocity from the plane and the wind.

[00344] The App may release the fire retardant bomb under Bluetooth control, for example. The pilot may release a safety catch just before the bomb is released.

[00345] Smaller aircraft may be used in the system for surveillance to spot fires and then to call in heavy water bombers. These planes may be fitted with sensors to detect fires and film the terrain which is then uploaded to the system. If a possible fire is detected, either by the pilot or observer, the instruments on the plane or from analysis of the system, then the plane should be instructed to inspect the location of the fire. The videos will map the fire location and size and measure other relevant information such as wind speed and direction, temperature and humidity. This information will enable a response to be devised and scheduled by the system.

[00346] Additional drone functions could be implemented as follows.

[00347] A drone may be instructed by the system to fly high to ensure that communications can be restored in an area for emergency workers and maybe for the emergency App and SMS for civilians in the emergency zone, as text messages for the emergency App and as SMS messages use a limited bandwidth.

[00348] Drones may provide communications to lost hikers and provide food, water, medicine, external batteries etc. Drones at sea may provide buoyancy vests and clean water etc. [00349] Drones and crop dusters may be used as a resource by the system to light fires to create containment lines in inaccessible areas or where these containment lines are needed very quickly. For example, strategic containment might be at the top of ridges. Burning the top half of the ridge on either side of the ridge is likely to contain most fires. The ridge may be chosen by the system based on data associated with minimising the risk of erosion if there is heavy rainfall.

[00350] Crop dusters may be used as a resource to spray a flammable chemical and set the area sprayed alight when they are out of range. Drones with fire suppressors may be used by the system to stop the fire from spreading.

[00351] An example Dynamic Fire Attack Scheduling System is now described.

[00352] The system may monitor and analyse all potential fires and their locations and prioritize them into priorities 1-10. The potential fires can either be inspected by resources to see if they are a fire or attacked with water or a fire control device in anticipation that they are a fire.

[00353] The system may monitor and analyse all identified fires and their locations, categorize by size, determine what response is needed, the probability of success of various responses and prioritize the responses to the fires into priorities 1-10.

[00354] The system may have a database of the available resources, such as planes, helicopters and drones for response, where these resources are located, what their fire attack capabilities are, and their readiness, whether they can attack a fire or whether they have to return to reload.

[00355] The following Scheduling examples are provided.

[00356] The scheduling problem may not be very complex when there are more aerial resources (planes, helicopters and drones) than fires. For example, lightning strikes may be evaluated by the system for data collection purposes and then the fire is attacked. For example, the site may be inspected again after the response and a decision may be made by the system to respond again or a determination made that the fire is extinguished and no further response is required.

[00357] As more fires happen, then the first response determined by the system may be to instruct more aerial resources to fight them and attack all fires. Strategies to improve the efficiency of the response may include: i) in addition, making the attacks more accurate and effective may reduce the number of attacks required to extinguish a fire; ii) accuracy of the fire location may assist in the accuracy, speed and effectiveness of the attack, iii) assistance with aiming the attack by calculating the best direction and height of the attack and telling the pilot exactly when to release the attack; iv) enabling equipment to be easily attached to planes, helicopters and drones to increase the number of aerial resources with minimal cost.

[00358] The aerial resources need to be classified as to their size and likely effectiveness on different sized fires and different fire types: different types of drones, helicopters, light planes and purpose built tankers. This will be based on prior experience. As these planes are involved in attacking different types of fires in different locations, and as the results are recorded, this table will be update and optimized.

[00359] In extreme or catastrophic fire weather, the number of fires is likely to exceed the aerial resources. The system may make determinations about which fires get priority based on the following. i) Some fires that are close to towns could be extinguished by ground based fire services if they are informed as soon as the fire location is identified. This will free aerial resources. ii) Multiple fires may be extinguished by one aerial resource on one flight. iii) Drones with limited fire fighting capacity may not be sent to a large fire, although a swarm of drones may be sent - the required response should be matched to the aerial resources available. iv) The current location of the fire and the wind direction may be used by the system to determine whether this is likely to drive the fire towards a town or key infrastructure which should be defended. v) Based on the terrain, the undergrowth and trees, the weather conditions and wind, the system may estimate the damage the fire will do if it is not attacked. vi) The system may look at whether containment lines have been built that will enable the fire in its current direction to likely be contained within the containment lines. vii) The system may determine how many resources and which resources are needed to attack the fire. viii) The system may determine what the opportunity cost is to attack the fire. This may include the size of the resource and the time used by the resource to e.g. fly to the fire, attack it, return to be refilled and take off again. For example, it may be most efficient to continue a resource on a fire attack rather than rerouting a resource especially if the resource is about to attack a fire. ix) The system may determine the cost of not attacking the fire. This may be based on whether there is any vulnerable infrastructure likely to be burned as well as other costs. x) The following Software Structure may be used by the system.

[00360] The system may store the data on one or more databases accessible by the server for the system to schedule emergency responses. The data may be held in a single database where someone (or an organization) is the custodian of that data. These databases may be connected together with software code to produce complex answers.

[00361] The information for a decision about fire attack priority for a particular location struck by lightning observed by sensing towers may ask the for the following information.

[00362] The weather conditions database may include the BOM database but may also have more specific information e.g. from a lightning sensing tower and from a drone. The database may present the most accurate and timely information known to the database with simplified data (in mathematical form) stating the weather conditions including the temperature, humidity and wind direction and speed.

[00363] The physical terrain database may be used by the system to respond with simplified data (in mathematical form) indicating, for example, that the terrain at that point is a rocky hilltop and that in 35 degrees on either side of the wind direction vector there is no infrastructure for 5 miles and then there is a defendable road. The database may estimate, for example, that the fire would take 8 hours to reach the road.

[00364] The vegetation database may be used by the system to respond with simplified data (in mathematical form) that the vegetation at that place has no trees and just a few shrubs.

[00365] The lightning strike database is provided with the terrain, weather and vegetation information and the system may respond, for example, that the probability of the lightning strike starting a fire in that location in those conditions is 20% and that the best time to observe a fire is in 15 minutes.

[00366] Human safety database may be used by the system to report that the area is inaccessible, for example, and with the lowest probability of a human being there and that a drone will have a 90% chance of seeing a person if there is one. [00367] The aerial resource database may be used by the system to report that a drone armed with a 25kg C02 bomb is available 10 minutes away, for example.

[00368] Th system decision may then result in a response to send the armed drone to leave in 5 minutes.

[00369] The drone sensors may pick up visual and infrared radiation indicating a fire and film the fire from all directions. The drone may transmit the photos to the fire attack database together with details of where the photos of the fire were taken from. The drone independently transmits accurate weather data to the weather conditions database.

[00370] The system may use the fire attack database to calculate the direction, height and speed of the attack and the drone receives these instructions from the server in a response.

The drone then films the aftermath of the attack and sends this back to the fire attack database. The system uses the fire attack database to calculate that the fire has been extinguished with an 80% probability and submits the fire location to the surveillance database.

[00371] The system may use the surveillance database to schedule for larger airborne drones to fly directly over the fire location in the next hour, 2 hours, 4 hours and 12 hours to see if there is any evidence of fire.

[00372] List of databases and system elements may include, for example, the following.

[00373] Communications system that will enable communications across multiple different communication technologies: internet, cell phone, messaging, satellite etc

[00374] Communication receiving unit that receives communications and send these communications in real time to the correct database.

[00375] Communications sending unit that will send information to resources via whatever channel is available and if there are communications bottlenecks prioritize the bottlenecks based on the priority of the signal

[00376] Terrain includes infrastructure, landing strips, condition of the landing strips (is the ground suitable for landing, are gates open and stock off the strip etc), is there water on the strip to reload the plane, airports, places where planes can skim lakes and the sea to load up with water. [00377] This database may describe infrastructure, critical infrastructure and the defensibility of infrastructure.

[00378] There may be a database of infrastructure that needs to be upgraded to ensure that it is resilient and can be defended.

[00379] Interrelated infrastructure data may be stored.

[00380] Electricity generation needs communications to reconnect to the network and electricity to start the generation equipment. These facilities should be available in an emergency to ensure that the grid can be restarted quickly.

[00381] Production of C02 bombs may happen locally as they have a short half-life. Tanks of C02 and other materials used in the process should be in the towns at the start of the fire season. It is essential that power to is available and that there are communications so that the delivery of the bombs can be achieved. Power can come from mains, solar, a battery and other forms of energy storage.

[00382] Country town are most at risk from electricity failure and significant energy storage should be used both to stabilize the grid, reduce energy costs and ensure supplies of energy in an emergency.

[00383] For example, infrastructure that has poor defensibility will need to be made more defensible. For example, transmission lines should be strengthened. Another example is national forests. In times of extreme fire danger, access to people in the forests should be denied but people in the forest should be allowed out. Installation of a drawbridge like structure will enable the weight of the car to push the drawbridge down for exit, but will stop new entrants.

[00384] For example, monitoring on national parks and key infrastructure will enable better intelligence. If a fire starts in the bush and the vehicles going into the bush are monitored, then there is a good chance that the perpetrators will be caught before they can light another fire.

[00385] For example, major roads should have trees thinned near the road with regular off season burning to ensure the road is open in a fire emergency and that the road can act as a fire containment line. [00386] Vegetation data may be stored, such as, for example, everything about vegetation at a site.

[00387] A weather conditions database may include flood information etc. A part of the database will list those regions likely to have high, very high or catastrophic fire danger and will pre-emptively suggest relocation of resources based on the location of the risk.

[00388] A lightning database may store all the potential lightning strikes and information about these strikes.

[00389] Lightning strike prioritization system may take data from the lightning database and other databases such as weather, terrain and vegetation and calculates the likelihood of a fire being started by the lightning strike and prioritizes inspection and/or attack on an ordered scale of 1-10 with priority 1 being higher than priority 2. This information may be automatically fed into the Scheduling system.

[00390] For example, the system may also describe the requirements of aerial or land resources required to inspect and/or attack the potential fire caused by the lightning strike. This may be on a non ordered list of 1 to 100.

[00391] For example, there may be more than one suitable resource to respond to the situation. For example, the response to a low priority lightning strike could be a large water bomber, a small water bomber, a drone swarm attack, a single drone attack or a drone inspection.

[00392] Potential Fires database may store all the information about fires other than those caused by lightning strikes and prioritizes them for inspection and/or attack on a scale of 1-10. This may then be fed automatically to the Scheduling system.

[00393] The system may estimate the rate or growth of the fire over time which may be calculated from or extrapolated from historical or experimental data based on similar terrain, vegetation, and weather, particularly humidity, temperature and wind.

[00394] The system may also estimate the damage a particular fire may cause if not responded to. This may be calculated from or extrapolated from the proximity of infrastructure and property, and historical, experimental and/or computer simulations of fires. [00395] The system may also take into account the probability of a supervening event, such as change of wind direction, rain etc. For example, if there is a high probability of rain in one area and not another area, if a fire must be left, it may be better to leave the fire in the area likely to receive rain.

[00396] The system may also describe the requirements of aerial or land resources required to inspect and/or attack the potential fire caused by the lightning strike. This may be on a non ordered list of 1 to 100.

[00397] For example, there may be more than one suitable resource to respond to the situation. For example, the response to a low priority lightning strike could be a large water bomber for a larger fire, a number of small water bombers for a larger fire, a drone swarm attack for a smaller fire, and a single drone attack for a small fire.

[00398] A fires database may store up-to-date information on all verified fires and prioritizes then on a scale of 1-10. This is about newly started fires and not about the established fires - although it will relate to spot fires near established fires. This information is fed into the Scheduling system.

[00399] An aerial resource database may store data concerning availability of resources, their capacity and where these resources are located. This may require the active redeployment of resources to areas at high risk of fire.

[00400] A land fire response resource database may be used to store data about which land based fire-fighting resources are available, their location and capacity.

[00401] An emergency response scheduling module may schedule the aerial and land resources to attack, suppress or extinguish the newly lit fires based on priorities submitted.

[00402] A surveillance database may store data on scheduled inspections of low priority sites and sites that have been responded to, to ensure that the fires are out.

[00403] An emergency response scheduling module may be part of the system with an associated database. This may be used by the system to perform multiple scheduling calculations simultaneously and compare these to find the optimal or satisfactory scheduling solutions, if available. [00404] Multiple scheduling scenarios may be run on the system and used to find efficient strategies.

[00405] For example, one strategy may be location: fast response requires response resources close to the fire and the movement of resources pre-emptively and in very fast response to the emergency.

[00406] Various scheduling approaches may be used by the system.

[00407] For example, the number of responses required (fires or potential fires) and the number of resources available. Divide the response types into categories as follows:

A) Responses requiring large bombers or a number of smaller aircraft;

B) Responses requiring a single small aircraft;

C) Responses which require one of more drones armed with C02 bombs;

D) Responses which can be done by surveillance drones.

[00408] The system may apply the resources to each of these categories. For example, if there are more large bombers than fires, then the large bombers can be used to bomb one or more smaller fires. If there are fewer large bombers, then the system can apply multiple small planes to these fires. If there are enough surveillance drones, one can be assigned by the system to each potential fire. If there are enough resources even in different categories, then all the resources can be assigned by the system simultaneously. Similarly, if there are more large bombers than fires requiring large bombers, the large bombers may be assigned by the system to put out multiple smaller fires.

[00409] If there are fewer resources than fires, then the system determines a suitable response for each fire whilst minimizing the elapsed time to respond to all fires, including returning to base and reloading and travelling to a new fire etc, whilst prioritizing the highest priority fires first. Reducing the number of false positives with potential fires will reduce this elapsed time by eliminating trips.

[00410] Multiple strategies may be run for different situations and an Artificial Intelligence (Al) system may be used to look at the input data and select the most likely strategies to produce an optimal or satisfactory outcome. Actual results from scheduling operations may be fed back into the system to improve the scheduling efficiency. [00411] The emergency response system may have an Artificial Intelligence (Al) and/or machine learning module or associated system that can communicate with the server. The server may have the Al and/or machine learning module formed as part of the server system.

[00412] This module may be trained using raw information that has been analysed either manually or by running different scenarios with the assistance of software. The raw information may include data received from one or more sensors in the emergency response system.

[00413] The machine learning module may be arranged to analyse machine learning data associated with the emergency situation. The server may then be arranged to generate and output the emergency response to the emergency situation to at least one of the resources based on the analysis of the machine learning data and the determination of how to respond to the emergency situation.

[00414] For example, the machine learning data may use one or more data sets for training purposes. For example, the training data sets may include data based on one or more of i) the analysis of undergrowth data and associated fuel content data, ii) analysis of image data related to potential lightning strikes, actual lightning strikes and no lightning strikes, iii) analysis of hypothetical emergency situation data and prioritised emergency response data, and iv) analysis of scheduled emergency response data in a hypothetical emergency situation.

[00415] For example, the analysis of the hypothetical emergency situation data, prioritised emergency response data, and scheduled emergency response data in a hypothetical emergency situation may be carried out manually and then entered into the machine learning module.

[00416] Further examples and scenarios are now provided.

[00417] Disclosed is a layered fire detection system whose purpose is to rapidly identify, accurately locate and then confirm fire ignitions so that a rapid response to suppress the fire can be initiated. In contrast to the current firefighting focus, which is on fighting established fires which are hard to contain and very difficult to extinguish, the new focus is now on detecting a potentially large number of small fires which are hard to detect and then put them out quickly and completely while they are still small and easy to put out. Having big helicopters quickly deliver small amounts of fire suppressant with pinpoint accuracy on small fires to extinguish them means that one helicopter can attend to many fires on the one sortie. Having the helicopter route optimized by software may increase the number of fires one helicopter can extinguish on the one sortie. The layered response may include some or all of the following components structured in such a way that deficiencies in one layer are compensated or corrected by features in another layer:

[00418] Also disclosed is a system of satellites (both geostationary and low earth orbit) that provide broad area initial detection of potential fire ignitions using a combination of visual and infrared sensors. Said satellites are structured to identify unique and defining characteristics of a fire ignition, including but not limited to the correlation to points of lightning strikes, the existence of particular heat or infrared signatures, the existence of particular smoke or air disturbances and other factors, either sensed through direct causal observation or identified through artificial intelligence deep neural nets that have been trained on prior large data sets.

As the satellites are potentially equipped with high resolution cameras and potentially with specific radar capabilities, the satellites are envisaged to also contribute ongoing data regarding the risk factors on the ground for fire ignitions, including but not limited to surface moisture, sub soil moisture, canopy fuel load, sub canopy and surface fuel load, local weather conditions and other factors. The role played by the system of satellites in the layered detection system is to identify potential ignition points, erring on the side of potentially high false positives and very low false negatives, such that other layers in the detection system can then filter these targets into confirmed ignitions using higher resolutions observations, potentially from multiple other layers.

[00419] Also disclosed is a system of observation (sensing) towers, located strategically across the area of control, whose purpose is to detect the location of fire ignitions or potential fire ignitions. The location can be determined by stereoscopic camera arrays and by triangulation, where such stereoscopic location and triangulation is achieved using techniques that include at least some of the following: a) Observations in the visual spectrum using arrays of cameras to identify characteristic visual signatures, such as lightning strikes, smoke and or flame; b) Observations in the infrared spectrum using arrays of infrared cameras to identify characteristic thermal signatures of fire ignitions against the background thermal pattern; and c) Observations of atmospheric radio static, or “sferics”, to identify the unique characteristic sferic pattern for lightning strikes of various types.

[00420] The role played by the system of observation (sensing) towers in the layered detection system is to identify accurately the location of possible ignitions and to confirm those that are in line of sight, detect smoke from an ignition if it is not in line of sight, whilst calculating the probability estimate of ignition for those that are not line of sight for which no smoke has been observed (which will reduce with time), and yet are potentially triggered by lightning. This layer supplements the detection information in the satellite layer by providing both an alternative mechanism to identify potential positives and also a mechanism to confirm exact locations of ignitions in many, but not all, circumstances, thereby rejecting false positives in the assessments.

[00421] Also disclosed is a system of overflight drones that are deployed during periods of high risk in particular areas, where such drones are equipped with arrays of visual and infrared cameras and where such drones may either be completely automated or may be remote controlled, and where such drones may fly at a variety of heights, including but not limited to low altitude drones operating in the sub 1,000m zone and or drones operating in the very high altitude zone in excess of 20,000m. The drones stream data to a server or servers, or may store data on board for later retrieval, where the said data is assembled to further identify and confirm fire ignitions and also to estimate the size and intensity of ignitions. The role played by this drone layer in the layered detection system is to positively confirm and then quantify fire ignitions, and to reject all false positives created by other layers. In particular this layer is structured so that it can confirm ignitions that are obscured in line of sight from the observation (sensing) tower layer (either because of terrain, such as valleys, or due to obstructions, such as vegetation or terrain, because of insufficient tower coverage in a region or because of factors such as high winds that may make the detection of smoke difficult). In addition it is envisaged that this layer may continuously generate additional data regarding ground level fire risk (moisture, sub soil moisture, fuel loads at canopy and sub canopy, local weather conditions and other factors) that can assist in the prioritisation, modelling and optimisation of suppression activity.

[00422] Also disclosed is a smartphone software application program (App) is provided as part of the system that enables the public to quickly report fires and associated hazards, such as road hazards, in near real time back to the server (either directly or via any other communication device in the communication network). Metadata provided by the App may contain GPS location information and other information captured by the smartphone. The smartphone application may also be used as a communication channel to provide information and communications from the Bureau of Meteorology, emergency services and police, for example and may provide data to emergency services as to where people are. The communication channel may also be used for other emergency applications, such as tracking bushwalkers so emergency services know where the bushwalkers are if an emergency arises.

[00423] Also disclosed is a system of suppression helicopters and other aerial craft together with associated infrastructure (forward bases, radar control, communications and logistics management) such that a subset of this fleet can be rapidly deployed in an optimal targeting pattern to suppress fire ignitions within a short time using various fire suppressant materials such as water, water with retardant, carbon dioxide pellets or bombs or other various retardant materials. The role of this layer is to be deployed in areas of high risk in advance (based on information largely from weather forecasters and from the satellite layer) and then to be used in targeted suppression of fire ignitions based on confirmed ignitions from the other combined layers, with the objective of the layer to proactively suppress fire ignitions before they evolve into larger and less manageable fires.

[00424] Also disclosed is a system of containment and support resources that are optimally deployed in areas of long-term high risk prior to any fire event such that these containment and support resources are designed such that they fulfil at least one of the following functions: a) Contain any fires that escape the ignition suppression functions identified above through the use of structured containment lines that may contain some or all of the following characteristics: a. A wide and graduated buffer zone where the fuel loads and vegetation ecology is maintained in such a way that fire transmission through the zone is severely attenuated, using some combination of indigenous fire control methods and newly researched mechanisms for both restricting fire ignition, restricting fire transmission, and suppressing wind turbulence; b. A capacity within the containment line to rapidly reinforce fire suppression activity by providing transport corridors (such as roads, helicopter landing sites, water tanks, drone refuelling or recharging sites and similar other resources); c. A system of maintain soil moisture during high risk periods using a combination of aerial spraying, local irrigation, specialized vegetation and encouraged animal, fungal or insect species that create or maintain soil moisture; d. Observation (sensing) towers that provide information to firefighters to fight the fire more efficiently, including information about the fire front and an ember tracking system that will tell firefighters where an ember is likely to fall or has fallen using stereoscopic arrays of visual and infrared sensors and triangulation to measure the path of individual and clusters of embers; b) Provide water or retardant resources that can be accessed by aerial or ground fleets such that there is an optimal supply of retardant (such as water) during a fire suppression event, and including such items as water tanks, dams, supply pipes, storage sheds, buffer zone clearance for access to the same; c) Ground based fire suppression resources, including but not limited to, fire trucks, automated fire trucks, automated fire hose targeting sites, automated and remoted control tankers and similar,

[00425] Innovative aspects of the containment lines herein described consist of: a) Software used for the design of containment lines b) Design of baffles within containment lines to disrupt the flow of air and reduce both transmission and ember attack; c) Ember tracking systems; d) The coordinated role played by containment lines in a multi layered suppression strategy

[00426] Many of the worst bushfires are started by lightning in inaccessible areas. Stopping or reducing lightning strikes may significantly reduce the bushfire threat. An innovative system is described that aims to stop and/or reduce the frequency and intensity of lightning by stabilizing the charge in clouds, to convert thunderstorms from dry lightning storms to wet lightning storms which may reduce the probability of ignitions from lightning strikes, and to stimulate lightning strikes to hit safe areas rather than places where lightning strikes are likely to ignite fires.

[00427] Further details of aspects of these layers are described below.

Aerial Fleet

[00428] The system described herein may also include a substantial aerial fleet, consisting particularly of large, general-purpose helicopters (certified to fly operations at night and in a wide variety of weather conditions), supplemented by fixed-wing water bombers.

[00429] The system envisages that contrary to existing practice, the aerial system described herein responds to small or relatively modest fires with overwhelming force with the aim of complete extinguishment or suppression. This is contrary to existing systems that more generally use aerial resources to retard existing large fires.

[00430] The characteristics of this layer are therefore some or all of the following: a) Sorties of resources towards many target locations, most of which are small; b) As targets are small and the objective is to extinguish quickly, targeting of suppression retardant materials such as water should be highly accurate (both to achieve the result and to maximize the effectiveness of the overall sortie that may involve many such targets) c) As the time-to-target is a critical factor in suppressing the fires before the fire grow to unmanageable proportions, the aerial resources should perform in a very wide envelope of conditions (night and day, high wind, low visibility);

[00431] This aerial layer may work in close consort with the other layers as it is may be necessary to have highly accurate target information, and directing a sortie to a false positive is a significant waste of resources.

[00432] Largely for the above reasons, the system described emphasises the use of helicopters where such helicopters have various innovative characteristics including but not limited to: a) A capacity to direct significant quantities of retardant to quickly and efficiently extinguish small fires whilst hovering at significant heights using a smart retractable hose that automatically adjusts the height of the nozzle assembly to avoid nearby obstacles and which directs the retardant onto the fire using self calibrating nozzle adjustments that determine the rate of fire, the dispersion of fire and the quantity of fire for the retardant delivery; b) The development of highly accurate 3D maps that will enable the helicopters to operate safely at night at a low enough altitude to effectively deliver sufficient quantities of the fire retardant to quickly extinguish the fire; c) A capacity to replenish retardant from a wide variety of locations (such as specialized tanks, dams, supply pipes and so on) through access to detailed 3D mapping of these sites by prior drone overflights that is then used within the helicopter to safely manoeuvre at night and in high wind or low visibility. d) Heuristic systems that may assist the helicopter crews to optimize the fire attack strategies for particular fires

[00433] Innovative aspects of this layer described within this patent include: a) An innovative system for targeting a water or retardant nozzle beneath a helicopter such that the helicopter can fly safely, can target accurately and can remain high enough that the downdraft does not reinforce the fire; b) By filming the fire with the helicopter above it at different levels, the system may calculate those conditions where it is possible to use the downdraft from the helicopter to help extinguish the fire by blowing the fire out, and mixing and cooling the air; c) An innovative system of coordinating multiple layers of information such that the efficiency of an aerial fleet of suppression aircraft can be deployed optimally; IQ

[00434] The helicopters may refuel by hovering over water close to a fire to reload fire retardant. The helicopters are arranged to drop fire retardant very accurately on the fire. Large durable water tanks that have been constructed in fire risk areas enable helicopters to fight fires with short transit time, increasing the effectiveness of the aerial response. Water tankers that have a hatch at the top of the water tank can also be used to recharge helicopters with fire suppressant close to a fire. Another advantage of helicopters is that less than 50% of fire retardant hits the target when dropped by fixed-wing aircraft. Helicopters can accurately drop a significantly higher proportion of their load, which is important for the rapid and effective suppression of small fires.

[00435] Helicopters may be fitted with lights and multiple sensors including heat sensors to enable the crew of the helicopter to inspect the site of a fire before, during and after suppression to better aim the fire retardant and to verify that the fire has been extinguished. These sensors can also be lowered from a hovering helicopter to enable close up inspection of the fires. The sensors lowered from a helicopter may be attached to the smart nozzle system or may be separate. This may mean that the fire location may not need to be inspected by another device, such as an inspection drone.

[00436] Drones may be used to help the pilot aim the fire suppressant drop by giving the pilot visual and other information about the fire. The drone used for aiming can also assess the effectiveness of the fire suppression activity. Additional aiming information can come from small single use sensors which may be dropped near the fire to provide suppression aiming information and provide evaluation information about the effectiveness of the suppression activity.

[00437] Sound in the 30-60 Hz band can be used to extinguish fires. In addition to lights and sensors to help pilots observe and extinguish ignitions, helicopters can be fitted with powerful speakers in the 30-60 Hz range that can be focused on the fire to extinguish or help extinguish the fire. These speakers can also be lowered so that they are closer to the fire.

[00438] Also described is a Smart Fire Extinguishing Nozzle System.

[00439] Helicopters may drop their fire suppressant from a low altitude to ensure pinpoint accuracy. As they drop in height, the downdraft from the rotors may increase the fire and may cause embers to spread. However, after wind from a helicopter reaches a certain speed, the wind may assist to blow out the flames. [00440] Some fires may start when lightning hits a tree. The top part of the tree may explode.

If the tree is hollow, a fire may start inside the tree. Bombing the surrounding area with water may not put out the fire. One way to put out the fire is to put a hose inside the tree and fill the inside of the tree with water. However, as it is difficult to manoeuvre the hose into the tree by moving the helicopter, a hose is described that can manoeuvre itself and lower itself into the tree to extinguish the fire inside the tree.

[00441] Similarly, a fire may start in a hollow in the tree that can only be accessed horizontally, so the hose may be configured to spray horizontally.

[00442] Water operates to suppress fires by turning into steam which reduced the available oxygen for the fire, by cooling the fire, by wetting unburnt fuel which makes ignition of unburnt fuel less likely and the speed of the water can end up blowing the flame away from the burning fuel. Spraying water instead of dropping water may reduce the amount of water needed to extinguish a fire and may mean that the helicopter can attack more fires in the one sortie. A fine spray that quickly turns to steam excluding oxygen, cooling the fire and the air and increasing the air humidity which reduces the propagation speed of the fire. It may be that wetting the area in front of the fire to stop or slow the spread of the fire, and then extinguishing the fire, may be an optimal strategy in some fire conditions.

[00443] Figures 19A to 19F show examples of a fire-fighting device in the form of a nozzle system.

[00444] In Figure 19A, a nozzle system 1901 is provided. The system has a fluid inlet pipe 1903 for providing fluid (e.g. water) to the nozzle system and a steel support cable 1905. The pipe 1903 is in fluid connection with a reservoir 1907 to enable fluid to be provided to the reservoir 1907. Fluid from the reservoir is provided to multiple fluid channels (1909A - 1909B) that are in fluid connection with the reservoir. It will be understood that only two channels are shown in Figure 19A for brevity’s sake, but that there are channels for each of the nozzle outlets. Each channel has a valve (1911) and a nozzle outlet (1913). The valve is in fluid connection with the reservoir and the nozzle outlet. The valves may be, for example, a solenoid valve and are used to regulate fluid flow. Alternatively, the valves may be a gate valve opened and closed in increments by a leadscrew assembly. (A leadscrew is also known as a power screw or translation screw). The valves are connected to a control system 1915 via a wireless medium, such as Bluetooth for example. The control system can be used by an operator to control which of one or more of the valves open and close to control the fluid coming out of the respective nozzle outlet. [00445] Figure 19B shows a nozzle outlet control system 1917. An aperture 1919 is formed between two opposing members (1921A, 1921B) on the nozzle outlet 1913. A regulating pin 1923 is adjustable relative to the aperture to change the size of the output of the aperture.

[00446] Figure 19C shows a solenoid operated valve 1925 for regulating fluid flow by moving the regulating pin 1923. Alternatively, the valves may be a gate valve opened and closed in increments by a leadscrew assembly to adjust the rate of flow and the type of flow, ranging from a mist to a fast moving stream.

[00447] Figure 19D shows the aperture 1919 in a completely shut state, with an opening force 1927 being applied to the regulating pin by the solenoid operated valve 1925 to move it to an open state, as shown in Figure 19E. A closing force 1929 applied to the regulating pin by the solenoid operated valve 1925 or a leadscrew assembly then causes the aperture to move towards a closed state.

[00448] Figure 19F shows a fire-fighting system assembly 1931 with a nozzle system 1901. Attached to the nozzle system are sensors 1933 for detecting heat, smoke, fire etc. Attached to the nozzle system are one or more speakers 1935 that operate at 30-60 Hz. Attached to the nozzle system is an anti-snagging cover 1937. Attached to the nozzle system is a light array 1939 to aid visibility for the users of the assembly.

[00449] A smart fire extinguishing nozzle system is described that has some or all of the following components:

[00450] The nozzle head may have a range of sensors attached including video cameras and infrared sensors to detect hot spots, and accelerometers and other devices used for positioning and controlling the nozzle head, the attack strategy of the helicopter, and the attack strategy of any unmanned autonomous vehicles (drones) which are being used to suppress the fire. The pilot and crew of the helicopter may have full access to the information from the sensors on the nozzle. The information supplied to the pilot and crew may include a 360 degree horizontal and vertical view from the sensor, together with a view of the nozzle from the helicopter. This may help the pilot and/or crew aim the nozzle. One major advantage of this is that the helicopter crew may know if there are any remaining hot spots and this may lessen the need for a drone to inspect the fire after the helicopter has attacked the fire.

[00451] The nozzle system may be raised or lowered from the helicopter. When the nozzle system is to be raised, the water supply may be cut off in the helicopter and the pipe is emptied. [00452] The nozzle system may be configured to minimize the risk of getting snagged by having nothing that juts out or is easily snagged. The winch is supported by a strong cable, such as a steel cable attached to a fast winch. The nozzle system may be designed to retract quickly if objects in its path are detected by quickly raising the cable. This retraction feature may operate automatically when sensors on the nozzle detect an obstacle and send a message to a fast winch attached to the helicopter to raise the nozzle. In addition, a message may be sent to the helicopter pilot requesting the pilot to raise the helicopter. The nozzle system may be jettisoned if it gets snagged.

[00453] The pressure in the hose may be regulated by the length of the hose which may determine the head of fire suppressant and therefore the pressure. Alternatively, pressure can be supplied by a pump at the nozzle, in the helicopter or both. It is important that a high flow of fire suppressant be achieved so that the fire can be extinguished quickly. The weight of the water may help stabilize the nozzle head. If further stabilizing is required, water pouches can be added on the hose slightly above the nozzle head.

[00454] The nozzle system may have a number of nozzles that can be independently opened or closed to provide gentle sprays using limited water to a large focused flow that may be used to fill a hollow burning tree.

[00455] The nozzles on the nozzle system can be moved relative to each other to enable a very large range of water flows which can be optimized for each individual fire. This is analogous to an adjustable shower head or an adjustable watering wand.

[00456] The nozzle system may be maneuvered by using nozzles to expel high pressure water horizontally. In addition, engines may be located on the nozzle head and may be used to increase the water pressure and/or drive fans. Compressed air may also be used to maneuver the nozzle. The nozzle may also have the capacity to self-stabilize using feedback mechanisms.

[00457] The nozzle may also have a speaker or speakers attached to it to enable the fire to be fought both by water and by sonic waves. Also, chemical smoke detectors and strong lights may be placed on the smart nozzle. For example, LED lights may be used which do not generate significant heat. This may enable pilots to see what is happening and for video to be captured and used by the learning system. [00458] The wind profile of the nozzle should be minimized to enable simpler operations in high wind conditions.

[00459] The nozzle may be able to adjust itself to cover the fire and immediate surrounding area with the optimal water drop size and the optimal water flow rate for that particular fire.

[00460] The smart fire extinguishing nozzle may also have an Al informed control system that can be operated by the helicopter crew.

[00461] Also described is an Al based smart fire extinguishing nozzle control system.

[00462] There are a large number of variables which include: wind speed, temperature, humidity, topography, quantity and type of fuel, dryness of fuel, type of fire (is the fire on open ground, in the centre of a tree that can only be reached from the top of the tree etc.) and so on. There are many variables for the helicopter crew to use to extinguish the fire including: amount of fire suppressant used, different fire suppression application methods (ranging from dropping the fire suppressant from different heights to spraying it very close to the fire using the intelligent nozzle), use of sound to extinguish or help extinguish the sound, use of the downdraft from the helicopters to blow out the flames and cooling and mixing the air and so on.

[00463] A manual control system may be created and used to model the nozzle system extinguishing fires using the small fire propagation model. Algorithms using machine learning algorithms may optimize the nozzle system performance and simplify the use of the nozzle system by helicopter crews using data collected from simulated and actual use of the nozzle system to extinguish fires. This system may suggest a fire suppressing strategy to the helicopter crew based on the fire conditions and the previous experience which has been encoded by machine learning into the system. This may include wetting the surrounding areas first to stop or slow the fire spreading as well as other variables such as the fire suppressant application method, use of the downdraft to blow out and cool the fire, use of sonics to extinguish the fire and so on.

[00464] The sensors (video cameras, lights, infrared sensors, chemical smoke detectors etc.) that are on the smart nozzle may be lowered to help pilots aim their water from their helicopter and more importantly allow them to observe post drop the effectiveness of the drop to ensure that the fire is out. This may be the most efficient way to extinguish the fires rather than hoping the fire is extinguished and the sending a drone to inspect and then sending back the helicopter is there is still a fire. Observation (sensing) towers

[00465] Various sensing towers are described herein with reference to Figures 18A to 18D. Also disclosed are sensing units with sensing towers.

[00466] Example sensing towers may be located in inaccessible places may be constructed using a helicopter to lift the tower into place.

[00467] Example sensing towers may be located close to roads and may be brought to the site in trucks.

[00468] Referring to Figure 18A and 18B, an example of a sensing tower being assembled on the ground is now described.

[00469] A foundation 1801 is provided in the ground. A lower base plate 1803 is attached to the foundation using a number of bolts 1805. A pivot point 1807 (e.g. a hinge) pivotally connects the lower base plate to an upper base plate 1809. A support element 1811 is positioned to support the sensing tower when on the ground. Small jacks 1813 are positioned for the tower section to be held in place for assembly. Once lifted into place, the tower may be bolted or welded at weld points 1815 to secure the column to the lower base plate.

[00470] Referring to Figure 18C, an example of how the sensing tower may be raised is provided. A cable 1817 may be attached at one end to a winch 1819 and the other end to a removable lifting sleeve 1821 placed around the column of the sensing tower. Also provided is a telescopic hydraulic jack 1823 with a telescopic lifting arm 1825 attached to the lifting sleeve. By using the winch and jack, the sensing tower can be raised into position.

[00471] Therefore, the sensing towers may be assembled and raised using hydraulic rams or cables, or both.

[00472] Sensor arrays of the sensing towers may be positioned on a drive system 1827, such as a rack and pinion, to enable the sensor arrays to be raised and lowered for maintenance.

[00473] An example sensing tower in Figure 18D has an observation platform 1829 above the two sensor arrays (1831A, 1831 B). In this example, the sensing tower is constructed from sections 1833. Electronic systems and back-up generators may be provided inside fireproof boxes 1835. Solar panel power systems 1837 may be provided. [00474] Figure 18E shows an example of an observation platform 1829 on a sensing tower.

The platform is bolted to the tower using bolts 1839. The platform has a safety rail 1841 and a floor 1843. A ladder 1845 is provided to enable people to reach the platform. A pulley 1847 is used to enable lifting up of equipment in a lifting harness 1849. A trapdoor 1851 is also provided to give access to the platform via the ladder.

[00475] The sensing towers may include cell communication modules, such as 4G/5G communication modules positioned on the sensing towers to enable communications to devices between the towers and people who are hiking and have access to suitable communication equipment (e.g. via their mobile device and a smartphone software application program). This enables the ability to report accidents and also provide the system with information if there are information gaps. The 4G/5G communication modules may also provide communications in emergency situations.

[00476] The sensing towers may also collect data for the Bureau of Meteorology, emergency services and police and provide this via the server (either directly or via any other communication device in the communication network).

[00477] The sensing towers may operate 24 hours a day, 7 days a week.

[00478] The sensing towers may operate with lightning storms above and have suitable lightning protection.

[00479] The sensing towers are arranged to operate automatically without human intervention.

[00480] Two, three or more sensing towers may be arranged to be in sight of each other to enable the system to use triangulation measurements to accurately determine the location of identified events (e.g. lightning strikes, fires, smoke etc.).

[00481] The sensing towers may be configured in an approximate hexagonal shape to provide improved triangulation measurements. This configuration of the sensing towers may depend on the topography of the land.

[00482] The sensing towers may be configured to be about 10 kilometres apart, for example. The distance that the sensing towers are separated may be calculated based on the pixels in one or more of the cameras in the sensing towers. In the tower camera array, there may be a short-range camera and a long-range camera, which may have a narrower angle of vision.

[00483] One or more of the sensing towers may use one or more spectroscopic cameras to measure the distance to fires within, for example, 5 kilometres of the sensing tower using the short-range cameras.

[00484] Triangulation methods may be used between the sensing towers using the long-range cameras.

[00485] One or more of the cameras on the sensing towers may pointed towards clouds in the sky to enable the system to capture and analyse images to resolve the location in the cloud where the lightning started, as well as the location where the lightning hit. The system may use the captured images to measure the path of the lightning. The determined path of the lightning may be used by the system to classify the lightning type and/or lightning shape based on the captured images. The path of lightning, lightning type (e.g. polarity) and/or lightning shape may be used to determine a probability of ignition of a fire. For example, an algorithm and/or machine learning system may be used to calculate the probability of ignition of a fire. The duration of the lightning may also be measured to give a measure of the energy flowing. That is, the longer the duration, the more energy may flow and the higher the probability of a lightning ignition. This may be factored into the algorithm and/or machine learning system to calculate the probability of ignition of a fire.

[00486] Weather information (temperature, wind, humidity etc.) may also be obtained at the sensing towers. The weather information may be used as part of the algorithm and/or machine learning system to calculate the probability of ignition of a fire during lightning.

[00487] The system may use the images captured by the cameras to measure the height of the clouds from the ground. The sensing tower height may be up to 35 metres to enable cameras (and other sensors) positioned thereon to see over obstacles, such as close by trees. Therefore, the sensing tower may be used by the system to locate where the lightning came from in the clouds and where the lightning struck the earth.

[00488] The cameras used to photograph lightning may need to operate in under 1 millisecond. To maximize accuracy, high resolution cameras may be used, which may produce a lot of image data. A processor may be placed in close proximity to the camera to enable the camera to process the image data in real time. The processor may be programmed to quickly detect changes between captured images. If there is no change, there is no lightning, and the processor may delete most images, except for an image in an agreed timeframe such as, for example, 1 image per 10 seconds.

[00489] The sensing towers may be used to measure those areas which are not under direct observation. Smaller towers may be used between the standard sensing towers to increase the area under direct observation. If lightning only strikes higher terrain, then sensing towers that observe the higher terrain will observe the location of the great majority of lightning strikes.

Data describing the topography of lightning strikes will inform the siting of the sensing towers.

[00490] The sensing towers may be used to measure the polarity of the lightning strike (positive or negative lightning). The sensing towers may have radio direction finding equipment to obtain extra information about lightning strikes.

[00491] The sensing towers may have a physical locating device positioned on top of the tower to enable the exact measurement of the distance between towers. For example, a small silver (shiny) spherical object may be placed on top of the sensing tower to enable the sensing towers to locate each other.

[00492] If there is atmospheric distortion of visual information captured by the cameras, this can be corrected in real time because the system knows where the pixels of the shiny ball should be and can recognize where the shiny ball pixels are, and the system can correct the location by shifting the pixels to where they should be.

[00493] Markings may be added to the sensing towers to enable a drone to determine how far it is from the sensing tower by measuring the size of the markings. The markings may be, for example, a white sign with a black square on it that is of known dimensions. For example, the marking may be a QR code, which may provide additional information.

[00494] A drone may be able to use information from several sensing towers to enable the drone to measure the distance from the sensing towers and triangulate its exact position, based on the known location of the sensing towers. Accurate GPS may also be used provide the same information.

[00495] Soil moisture meters may be placed into the soil around the sensing towers to measure soil moisture. The soil moisture may be measured from planes and drones using ground penetrating radar. The readings from the soil moisture meters near the sensing towers may be used to calibrate the ground radar information from the planes and drones.

[00496] The sensing towers may collect information about the energy in the air using electrometers.

[00497] Drones are described for use with the herein described system.

[00498] A variety of drones may form the drone layer, with different purposes and functions. These may include some or all of the following types: a) A very high altitude drone that flies above the jet stream in the vicinity of 20,000 m and which uses visual and infrared cameras to supplement the satellite layer; b) Inspection drones that fly at low altitude, typically a VTOL system, such that the drones can take off and land automatically and where they possess arrays of cameras used for detailed mapping and inspection of terrain and fire sites; c) Suppression drones which may be used for suppression activity by dropping various retardants at small fire ignition sites, and which may be VTOL or fixed wing or may be air launched. Some drones may be powered gliders, whist other drones may be UAV helicopters;

[00499] The system may use different drone functions using different drone types for different purposes, such as, for example mapping drones, inspection drones, attack drones, high altitude surveillance drones etc.

[00500] Mapping drones with cameras and equipment, as described herein. The mapping drones may develop very detailed 3D maps for use by the helicopters to fly in very dark nights. The mapping drones may use a spectroscopic camera array to map the terrain in 3D very accurately. This information may be provided (e.g. directly or via the server/network) to the helicopter navigation systems to enable the helicopter to navigate to a fire at safe operating altitude (1000 ft to the ground) and then drop down to bomb the fire with pinpoint accuracy.

[00501] Inspection drones may be used to determine if a lightning strike has caused ignition. These drones may be relatively small and light weight and may be lightning proof. The drones can be fitted with a variety of sensors including cameras for the visual and infrared spectrum, chemical smoke detectors and a variety of lightning detectors.

[00502] The inspection drones may be trucked to near the lightning storm or flown to the lightning storm and dropped from a plane or helicopter. When dropped from a plane or helicopter, they may be able to stay flying longer as the energy to take off may be supplied by a lifting helicopter or plane, and the helicopter or plane may often be able to get closer to the lightning strike locations in inaccessible areas than a truck would. Planes and helicopters can fly quickly to the storm, so the inspection drones do not have to travel long distances.

Inspection drones dropped from planes and helicopters may therefore be slower than other inspection drones. Drones may be airborne for the duration of the lightning strikes and travel around after the lightning strikes have stopped, continuing to inspect strike locations. Inspection drones may be airborne for a long time and be able to fly low to the ground for detailed observation of lightning strike sites. On dark nights they may utilize lights or flash photography for visual inspections. Drones may communicate in real time with reasonably high bandwidths. Drones and the equipment inside them are configured to survive a lightning strike. Inspection drones may be small. Inspection drones may be designed to help helicopters aim their retardant drops e.g. by measuring the wind around the fire and showing the helicopter pilots where the fire retardant is landing. The surveillance drones may be dropped from the rear door of a helicopter or plane.

[00503] Inspection drones may have wings so that they can stay aloft for a long time.

[00504] The inspection drones may be launched from the tail ramp of a plane by attaching the drone to an arm facing in the direction that the plane is flying that can be extended below the plane in non-turbulent air. The engines start and the drone is then dropped by the plane. The arm is retracted, and another drone is attached to the arm.

[00505] For launch safety, the drones may be designed to fly above the speed that the plane is flying at.

[00506] The inspection drones may be vertical take-off and landing (VTOL) drones so they can hover over a lightning strike. They may have lights to improve video recording or photography capabilities. They may have broadband connection to send the information back to the main system.

[00507] Attack drones may be used to attack any ignition as soon as it is detected, or the probability calculated is above a defined threshold. To be effective against a range of fires, the attack drone may be able to drop a reasonable quantity of effective fire retardant (e.g. 500kgs) with pinpoint accuracy on a fire of 1-5m2. Dropping drones via the rear door of a plane or helicopter may be easier than a side door. The slower and lower the drone is flying over the fire, the easier it may be to hit the target with pinpoint accuracy. [00508] A solution to overcome the power requirements to take off with heavy payloads (e.g. 500kgs) is to use a powered glider facing in the direction the plane is travelling that is dropped from underneath the wing of a short take-off and landing (STOL) plane. The powered glider glides until it is near the fire, and then slows down and manoeuvres itself over the fire travelling at a slow speed and accurately drops the retardant on the fire. The drone then has enough power to fly without the retardant to a place where it can be retrieved.

[00509] As it is not safe for drones to transport C02 or any sort of explosive device inside a helicopter or plane, the drones containing these materials would be outside the plane or helicopter, e.g. mounted under the wings of a STOL plane. These STOL planes may operate out of local landing strips. These STOL planes may have a rear door and may fly as slowly as 120 knots.

[00510] An attack drone may be a vertical take-off and landing (VTOL) drone or a drone that requires a landing strip. Drones with wings are usually much more energy efficient than drones without wings.

[00511] VTOL drones may hover over a fire and drop their fire suppressant load at low altitude to get a pinpoint accurate drop. Whereas, drones with wings that are not VTOL may not hover so their accuracy dropping fire suppressant may be lower than a VTOL drone.

[00512] A low cost drone option is preferred as some drones may be lost in thunderstorms.

[00513] The attack drones may be manufactured using carbon fibre with a fine copper mesh within the carbon fibre.

[00514] A Lightning Information Collection System is described for:

1. Gathering lightning related information from all existing sources.

2. Having attack drones and weather balloons in a cloud with instruments.

3. Using sensing towers to measure the path and the shape of the path between the cloud and the ground and the duration of the lightning strike to provide valuable information about the probability of ignition. Further, the polarity of the lightning may be used (positive lightning is much more likely to start a fire then negative lightning). Positive lightning is much less frequent than negative lightning, but it may strike a long way from the negative lightning clouds. Other information that can be collected is using electrometers to measure electric charge of the air, height of clouds above the ground etc. In addition to visual and magnetic information, radio atmospherics may be detected by sensors and used. Strategically placed receivers may be used to triangulate the location of lightning. This may be used as an additional way to measure the location of lightning strikes.

4. Using satellites to measure the extent and type of cloud, opacity and colour of the clouds, microwave radiometers with relatively large antennae can look at the moisture content of clouds.

5. Sending inspection drones to locate the exact position of lightning strikes to see if there is an ignition.

[00515] This information enables machine learning to be applied to predict lightning ignition outcomes to reduce the false positives and negatives. The sensing towers may always be on so that if a false negative happens and a fire starts, then it may be observed by the towers when e.g. it starts to generate smoke, or the flame becomes visible.

[00516] A heuristic system is described to pinpoint the location of lightning strikes and predict ignitions at the lightning strike locations.

[00517] The determination of the exact location of a lightning strike allows the efficient inspection of the lightning strike location, because the inspection drone does not need to search for the lightning strike location.

[00518] Measuring the path and the shape of the path between the cloud and the ground and the duration of the lightning strike may provide valuable information about the probability of ignition, as will the polarity of the lightning (positive lightning is much more likely to start a fire then negative lightning). Positive is much less frequent than negative lightning but it may strike a long way from the negative lightning clouds.

[00519] Cloud height from base of the cloud to the ground is also important as high clouds may produce rain that evaporates by the time it hits the ground. Dry lightning is therefore more likely with high clouds.

[00520] The heuristic system may use multiple sensors to locate the lightning strike location, use the exact position to develop a relationship between the lightning strikes and topography (lightning is thought more likely to strike high ground or high objects), send in a drone to photograph the location, and then use this information to modify the lightning strike location algorithm to improve accuracy of location. [00521] Lightning ignitions prediction algorithms may require multiple inputs including fuel loads, fuel and soil dryness at the strike location, weather conditions (temperature, wind, humidity etc.), topography, as well as all the information collected about the lightning strike, including its polarity.

[00522] The system may use a number of different algorithms based on different assumptions which are compared to the actual ignition results from observations by drones or by seeing smoke if there are too many strikes to be investigated within a short time frame. When enough data is collected, the system may use machine learning algorithms to predict the location of the lightning strike and the probability of ignition.

[00523] For example, if the weather conditions are mild, the fire risk is medium, and there is some moisture on the ground, then the chances of lightning strike ignition may be determined to be low and it may be that the most cost effective solution is simply to observe with the sensing towers and only respond if an ignition is detected by the towers. A different response would occur should the day be a catastrophic fire danger day.

[00524] Described is a heuristic system to predict where lightning is going to strike.

[00525] Thunderclouds have a life cycle as they build charge, mature and then decay. Therefore, there is a time in the cycle and a place or a number of places in the cloud where the system may attack the cloud to reduce the lightning threat by stopping or reducing the charge build up, stimulating lighting to hit safe areas and by discharging the lightning within the cloud.

[00526] Information about clouds includes: observations of the clouds from observation (sensing) towers so that the system can resolve where in the cloud the lightning came from as well as where the lightning struck, the path the lightning took, its duration, information from satellites, especially those satellites with microwave radiometers with relatively large antennae to measure the water distribution in the cloud, local weather conditions measured at the sensing towers, including cloud heights and colour, topography of the area, fuel loads and other factors that may give the probability of an ignition in a particular area, models of the structure and the charge within a cloud, and relevant information from weather models. Further information may come from weather balloons or drones with suitable instruments flying in the clouds. This model may teach the system what areas of a cloud bank pose a significant threat of fire and should be responded to. [00527] Described is a heuristic lightning analysis and response system in which the aims of the heuristic lightning analysis and response system are to:

1. Reduce the frequency of lightning.

2. Reduce the intensity of lightning so that lightning strikes are less likely to ignite fires. This system may be able to discharge enough charge to turn positive lightning strikes into negative lightning strikes.

3. Stimulate direct lightning strikes to areas where a fire is unlikely to start.

4. Turn dry lightning into lightning with rain which will reduce the chances of ignition. One of the reasons why dry lightning occurs is that the base of the clouds is high, and that rain happens but evaporates before it hits the ground. Increasing the amount of rain will increase the likelihood of rain striking the ground.

[00528] Stimulating lightning strikes to the ground may reduce the lightning threat by having lightning hit safe places on the ground where the lightning will not start a fire.

[00529] Lightning rods may be utilized in significant risk areas. For example, if there are clusters of lightning strikes in an area that might cause significant damage if a fire was ignited, lightning rods may be used, which are passive and always on and will not cause risk to people and property.

[00530] If there is no cloud to ground lightning, then there may be no fires. Therefore, one response method is to try to generate cloud to cloud (CC) and intracloud (IC) lightning to equalize the charge in a cloud. To stimulate intracloud lightning, a channel may be developed between the negatively and positively charged parts of the cumulonimbus or thunder cloud. This can be achieved by, for example:

1. Cloud seeding with a substance such as silver iodide may cause cold water in the cloud to condense into larger drops and turn into ice and fall as rain or hail. Rain may reduce the likelihood of ignitions. Additionally, the silver iodide can cause positively charged ice crystals to form together and get heavier and fall towards the negatively charge base of the cloud, causing intracloud lightning when they get close enough.

2. Firing an explosive into the cloud where it explodes and causes a plasma or ionized column of gas that may allow the charge at the top and bottom of the cloud to stabilize causing intracloud lightning.

3. Another way is to create a plasma between a point on the ground and the thundercloud which may cause the cloud to discharge and create lightning. This can be done by firing a rocket into the base of the cloud which drags a wire and/or creates an ionized column of gas that enable the charge to travel down to a safe place on the ground. The material in the solid fuel to create the ionization may be caesium salts (non radioactive caesium salts are only slightly toxic) and in liquid fuel is calcium chloride.

4. Suspend a wire within a cloud to cause a conductive channel to enable the charge built up in the cloud to discharge through the wire. The rolled wire can be attached to a balloon which rises to a place where there is a strong charge and then the balloon lets the wire unroll. The thickness of a cumulonimbus cloud could extend to over 30,000 feet. This approach may be restricted to thinner thunder clouds which are usually found in the more temperate zones.

5. Dropping an object that will fall through down the cloud and produce a column of ionized gas as it falls. The object would lose mass as it gives off ionized gas and would completely disintegrate before it hits the ground, so the risk of the object hitting someone would be avoided. An altimeter can be built into the object that will initiate the destruction of the object before it gets close to the ground. The ionized gasses would attract the positive charge at the top of the cloud so that the distance between the negative and positive parts of the cloud will reduce and when small enough, intracloud lightning would be initiated. Dropping a ball that will produce an ionized column may be low cost. If the interior of the cloud is very turbulent, by the time the object nears the negative charge at the bottom of the cloud, the channel might be dispersed, so it is best if the object can be dropped where there is a strong downdraft so that the column of ionized gases is created quickly.

6. Gliding, parachuting or using a rocket or similar means to place a device into the cloud at the right place and at a safe height to cause an ionizing column or a plasma to be created to discharge the cloud. The device may be fitted with a fuse that may only arm when it is in a safe position e.g. at a sufficient height that the plasma will not strike the ground and start a fire. Given the turbulence within a thundercloud, the faster the ionizing column or plasma is created, the more likely that conduction of the imbalanced charge in the cloud will enable the cloud to discharge its positive and negative charges.

[00531] The herein described system may collect information about what is going on inside the cloud using microwave antennae on satellites, and weather balloons and attack drones which are fitted with instruments that may provide information about the dynamic structure of the cloud. This information may be used to provide input to weather models that can simulate the situation inside the cloud.

[00532] Described is a heuristic system to recommend location and type of response to a calculated lightning threat. [00533] The heuristic system may recommend the location and type of response to a calculated lightning threat, and aims to answer the following questions:

[00534] When in some situations, any early intervention equalizing the charge of a cloud while it is building charge may reduce the charge and size of the cloud, and lessen the intensity of later lightning discharges.

[00535] Where this involves locating the areas of a cloud that are likely to cause problematic lightning. Positive lightning is far more dangerous than negative lightning, so predicting those areas that can generate positive lightning is valuable to know. There are ground areas where lightning strikes are unlikely to ignite fires, e.g. on rocky ridges. However, the situation is further complicated because the clouds will usually be moving relative to the ground.

[00536] What - what response is suggested and where and when this response should happen. Where may include a 3D location within a cloud.

[00537] Modelling may be applied based on captured data to see how best to discharge clouds e.g. by producing an ionized column within a cloud and how long it would last before it is dispersed.

[00538] Described is a lightning strike inspection scheduling system in which, the system may schedule inspections of an area based on the probability that a lightning strike will cause a fire in that area.

[00539] There may be 3 different kinds of lightning, as follows:

1. a normal negative lightning strike that usually lasts around 1ms. Less than 1% of these strikes can cause lightning;

2. A positive lightning strike. This will be more likely to cause lightning. These strikes should be prioritized for inspection over category 1 ; and

3. A continuous lightning strike that lasts much longer than 1ms and therefore carries a lot more energy. This type of lightning is likely to cause fires and all such strikes should be inspected.

[00540] Further, the following 4 situations may occur with lightning strikes:

1. A strike and no ignition;

2. A strike and there is ignition, but it goes out by itself; 3. A strike and it starts a fire which generates smoke, heat and the flames are visible within 5-20 minutes; and

4. A strike that hits a tree and causes the inside of the tree to start burning. This is usually caused by a positive lightning strike, with continuous current, which lasts a lot longer than normal lightning flashed that usually last around 1ms. A tree burning on the inside may be difficult to see from the outside, however this may be detected with heat sensors. One way to deal with this would be to send someone into the forest to cut the tree down and extinguish the fire. An alternative is to use the intelligent nozzle to flood the inside of the tree with fire suppressant. With the data collected by the herein described system, Al may be used to predict which lightning strikes need special attention.

[00541] One or more of the following factors may be taken into account by the system when scheduling inspections:

[00542] Positive lightning is more likely to cause fires than negative lightning;

[00543] Continuous current lightning is more likely to cause a fire than a much shorter lightning flash, hence the system measuring the duration of the lightning flash;

[00544] Where the lightning hits is also important. If the system determines that the location has a lot of fuel, its dry, and there is a strong hot wind with low humidity blowing the fire up a rise with low humidity, this may be determined to more likely cause a fire than an areas where there is little fuel, the ground is moist, there is low wind etc.;

[00545] Some lightning strikes may get hot but then self extinguish; and

[00546] A large number of lightning strikes may occur in an area and there may not be enough drones to get to the strike area in a defined response time, e.g. 30 minutes. Therefore, within 30 minutes, there may be some uninspected ignitions that turn into fires. These fires may be detected using the sensing towers and satellites. When the fire is detected, the system does not necessarily need to inspect the fire but may instead send communications to call in the helicopters and attack drones to attack the fire.

[00547] Example scenarios may include the following:

[00548] The longer a lightning strike is left without there being a visible fire or smoke, the lower the probability that the system will need to send instructions to have the location inspected. If after say 30 minutes there is no fire for a normal negative lightning strike, then the probability is that there is not going to be a fire. If after an hour nothing has happened, then the probability will be very low a fire has started. Either there was no ignition or the ignition self extinguished. A machine learning system may be used to predict which lightning strikes need to be investigated, and in what order.

[00549] In a lightning storm lasting 3 hours with say 500 lightning strikes, the drones may start inspecting the highest priority lightning strikes and work their way down. As soon as an ignition is observed, the helicopters may suppress the fire. An ignition might be observed by the sensing towers or the satellite without inspection. After a couple of hours, normal lightning strikes that have not turned into fires are unlikely to do so, and so the system may determine to have the drones inspect the effectiveness of the suppression activities, or maybe instruct drones to inspect lightning strikes from other storms. If there are no pressing other activities, then the drones may inspect the strikes to gather additional data for the machine learning algorithms.

[00550] Al and machine learning may be used to analyse multiple lightning strikes to prioritize inspections and at some elapsed time after the lightning strike, the system may not inspect the lightning strike as the probability will be below a defined threshold used to indicate that there is a fire. The system (e.g. sensing towers, drones etc.) may continue to observe the lightning strike location for fire and smoke.

[00551] The sensing towers may be used to measure rainfall at the towers. The towers may also measure ground moisture. The system may use aerial resources with ground penetrating radar to measure the moisture under the soil. The direct measurement of ground moisture by the towers may enable the accurate calibration of the measurements by ground penetrating radar. Visual images may be analysed to detect rain within the images. Thunderclouds may produce rain in some areas and dry lightning in other areas.

[00552] The system may have data showing when it rained as the system directly measures the rain at the sensing towers (and possibly elsewhere). Photos (or videos) of the rain may be stored showing what the rain looked like. The system can use all this information in a machine learning system to come up with an algorithm that can visually detect where it is raining and how much rain. If it is raining where there is a lightning strike then there will be a lower probability that the lightning strike has caused an ignition. [00553] The system may use small fire propagation models that may predict the fire growth based variables which include some or all of the following: weather conditions (wind speed, temperature, humidity), soil moisture, fuel load, fuel dryness, topography (fires like to burn uphill and the wind may not be able to get to the fire if it is sheltered from the wind) etc.

[00554] The small fire propagation model may be used for prioritizing which ignition to suppress with the ignitions based on a number of factors including the location (several fires might be close together and can be suppressed in one helicopter sortie) and the damage the fires will do whilst left unattended.

[00555] The fire propagation model may provide an effective and detailed fire suppression strategy for each fire. For example, a pilot may be provided with a map of the ignition which indicated where and how much fire suppressant should be dropped in what order in what place to best contain the fire.

[00556] Smart Fire Containment Lines are also described. Fires can propagate in several different ways which include:

1. Radiant heat from the fire. The most effective way to reduce radiant heat is to distance the hot fire and the containment line by removing the ground fuel and thinning the forest so that there is less fuel to burn so that the radiant temperature is reduced.

2. Along the ground. This can be the burning of fuel and dry grass. One thing to do is to remove as much ground cover as possible for as large a distance as possible to reduce fuel load - fires need fuel to burn. However, in some cases, fires can travel along bare ground by igniting grass and other roots. This effect can be reduced by causing wind turbulence at ground level e.g. by having a line of ploughed ground the air needs to traverse. The ground propagation can also be stopped by having a line of moist ground, which stops the ignition, creates steam which reduces the partial pressure of oxygen and cools the air. Other ways to create barriers to ground fire propagation are discussed below.

3. Canopy burns and moves to nearby canopies. This can be reduced by thinning the forest - reducing the number of trees by area so that there is more distance between the tree canopies. The probability of canopy ignition is reduced as the area of canopy is reduced: many embers will land on the ground rather than in a tree canopy. The width of the containment line will also help to reduce the canopy transmission as embers will only travel a limited distance which varied with wind speed.

4. Embers float up and are carried by the wind. One way to slow the spread of embers is to create wind turbulence or use natural topography to cause turbulence, as embers are likely to fall to the ground in turbulent wind. If the ground is moist where the embers land, many embers will self extinguish. They can be manually extinguished by hitting them with a shovel or by extinguishing then with water.

[00557] Fire containment lines are dangerous places. People run a risk of burns, smoke inhalation damage, suffocation from lack of oxygen and accidents which can be caused by falling trees and vehicular collisions which heavy smoke makes more likely. It is a good idea to minimize the number of people needed to successfully defend containment lines to reduce the risk to fire fighters.

[00558] In addition to stopping the propagation of fires, fire containment line features should have some or all of the following additional features:

1. Safe vehicular ingress and egress.

2. Access to water. This may be done by siting the containment line near to a lake, dam, river with reliable water, town water or by constructing appropriately sized, large, durable water tanks which can be used for the land based and aerial firefighting. These water tanks may be sited in places where they can be easily refilled e.g. by rain capture, pumping water from another source and by water tanker which means that there should be suitable road access. The water tanks may be in an open enough area so that helicopters can load up with fire suppressant from the tanks. The water tanks can have buried water pipes from the tank to the places where the fire will be fought from to enable fire fighting equipment to conveniently connect to the water in these places. These pipes can be gravity fed with pumps on the fire fighting equipment or in cases of important infrastructure, can be high pressure pipes with hydrants.

3. Fire shelters may be provided if there is a chance that firefighter egress could be cut off by the fire.

4. Siting containment lines in places where there are favourable conditions. Siting containment lines along or near to freeways, roads, railways, places where there is little vegetation, lines in forests where there are few trees, natural clearings in forests, places where there is little fuel naturally occurring (this may also reduce maintenance costs), naturally occurring high soil moisture content and other favourable topographic features such as rivers, creeks, gorges, ridges, rocky outcrops, cliffs, etc., can simplify and lower the cost of building containment lines.

5. Siting containment lines to protect valuable infrastructure, villages etc.

6. The containment lines may work in all weather conditions especially in any wind direction. [00559] Further, forests may be thinned, and bark and leaf litter may be collected. The bark and leaf litter may be processed in situ to make biochar. The increase in the value of the biochar may enable commercial exploitation of the forests while the forests are being rehabilitated.

[00560] Also described are baffles to reduce or stop the propagation of fires along the ground.

[00561] The baffle system is described with reference to Figures 17A to 17F.

[00562] Ground baffles may generate turbulence near the ground level to disturb the laminar flow, drop out low embers, and mix and cool the air. These baffles may be made of dirt, such as ploughed earth, irregular piles or earth, a trench dug with the soil heaped next to it and so on. Sandbags of various shapes may be used and filled with sand or soil. In addition, baffles may be made of rolled fire hose laid out in a zigzag and filled with water. If small sections of hose are used, and the ground is reasonably level, these small sections may be filled with water and air. If the ground is not level, filling entirely with water may be desired. A small fence able to withstand wind may be constructed in an irregular pattern to cause turbulence. These ground baffle designs may be modelled in different terrains, soil types, ground cover etc. so that the best baffle can be implemented at each location.

[00563] Preferably, as much moisture should be captured in the soil along the containment lines as this may slow the propagation of fires along the ground and assist in extinguishing embers. Earthworks can be used to create barriers to fire propagation along the ground, and these earthworks should also be used to collect water. Holes of say 10cms in diameter can be bored in places where water will flow or collect, and these holes will both aerate the soil and trap moisture. In addition, the ground can be deep ripped and this will both aerate the soil and collect water. This will stimulate the ecology in the soil, and this will further capture soil moisture. Native grasses may be planted on the ground as these grasses have evolved to grow in summer and they are usually moist and green. For example, Australian grasses will stop the soil drying out.

[00564] Baffles may be used to reduce the risk of embers breaching the containment line. Embers are picked up by the wind and transported in the direction of the wind. Embers will travel further if the wind is strong. Baffles are provided that withstand strong winds. The baffles may present a smaller wind profile in strong winds. [00565] In most situations, there may be baffles on either side of the containment line (the fire might come from either direction), and so embers that get around the first set of baffles may then encounter a second set of baffles.

[00566] In addition to the mechanical baffles, strategically planted evergreen trees with high water content leaves can act as ember baffles. These trees will take a long time to grow, but they may add an additional ember defence as they grow by increasing the turbulence around the baffles.

[00567] A variety of ember baffles can be used. These baffles may rotate like a windmill so that they face into the wind, or they may be stationary. One stationary design is a vertical baffle with a triangular cross section, with a thin galvanized steel plate fixed to the vertices of the triangle with multiple holes in the galvanized plate. Many other designs are possible. Turbulence may be induced both in the wind hitting the structure and in the wind moving around the structures. The height of the baffle at each location may be determined by the aerodynamic baffle modelling system using the shape of the baffle selected, the location of the baffles, the topography of the site and the likely height and trajectory of the embers.

[00568] The baffle array may create turbulence in every direction but may be optimized in the direction in which a fire is likely to come, which may be the usual prevailing wind direction in that area. However, the wind on catastrophic fire day might come from a different direction and this should be checked in the prior fire information fire data module. This may be determined by the aerodynamic baffle modelling system and the topography of the site.

[00569] Another example way to build a baffle is to have two poles erected vertically in the ground with a mesh between them. Covering the mesh is a flame resistant, fire resistant, UV resistant cloth that is looped over the mesh and connected to base of the mesh at the base of the mesh. This cloth has holes cut in it to allow the flow of air through the cloth and the mesh, creating turbulence both for wind as it goes through the cloth and for wind as it goes around, above and beneath the cloth. The baffles are set out in an array calculated by the Smart Containment Line Design System to ensure that the right degree of turbulence may be created for embers to drop safely to the ground in places where they will either self extinguish or be able to be extinguished easily e.g. by a roller or spray.

[00570] The baffles should preferably be:

1. Low cost to build;

2. Produce little waste in manufacture; 3. Can be mass produced from common materials in Australia, such as galvanized steel;

4. Have little maintenance;

5. Last a long time so they can be amortized over 30+ years;

6. Easy to assemble in situ by bolting standard components of the baffle together and then raising the baffle vertically using an extending hydraulic jack and/or winches; and

7. There may be 2 versions of baffles: Tall baffles which may be constructed from heavier materials, and medium baffles which can be constructed of light materials. The tall baffles may need to be guyed.

[00571] An example of a baffle system is shown in Figure 17A. The baffle system 1701 is constructed from two columns made of pipe sections (1703A, 1703B) bolted together and connected by a number of horizontal bars (1705A, 1705B). Baffles 1707 are placed over the horizontal bars while on the ground. The columns are then brought into an upright position by winches and/or hydraulic jacks. The baffles are held in place with small gravity operated clips. Two baffles are made out of one sheet of light galvanized steel by pressing the folding the sheet. The columns (1703A, 1703B) are secured in foundations (1709A, 1709B).

[00572] As shown in Figure 17B, the horizontal bars are secured to the columns by welding the horizontal bar to sleeves (1711A, 1711B) place over the columns. The sleeves are secured to the columns using a nut and bolt system (1713A, 1713B).

[00573] Figure 17C shows an example of how to manufacture a pair of baffles (1707A, 1707B) using a press. After the sheet is pressed, two baffles will be created, which are then folded as in Figure 17D to hang over a horizontal bar 1705B. The baffles have formed therein slots 1715 for gravity clips 1717 (see Figure 17D). The baffles are hung over the horizontal bars as shown in Figure 17D.

[00574] Figures 17E and 17F show examples of baffle systems and how they disrupt the flow of air 1719 to cause turbulence 1721. These provide examples of how to position the baffle systems to minimise the number of systems being used.

[00575] Smart Containment lines are designed to reduce the number of people needed to defend the line. The containment line should be sited in the most favourable location for containment. The containment line should be designed so that as much fire extinguishment can occur passively, e.g. by ground fires hitting a barrier and stopping, and by having a large percentage of the embers being made to fall in a defined safe area where they will not start a fire such as an area which has moist ground so they will self extinguish or can be extinguished by a vehicle pulling a roller and/or a spray. Sensing towers may be sited in proximity to the containment line to give fire fighters detailed fire intelligence. An ember tracking system may be used to track embers which may be blown over the containment line so that its landing position can be accurately determined for firefighters to quickly extinguish, whether from the land or air. The ember tracking may use visual and infrared spectroscopic camera arrays and triangulation to accurately track the route of the ember. The information collected by the sensing towers which may be recording all the fire activity, the fire fighting activity in the containment line and the information collected by the ember tracking system may be used to update and improve the containment line modelling software which may result in a smarter containment line design system.

[00576] There are many different factors that may determine the location of a containment line: topography, fuel load, proximity to assets that need to be defended etc. Soil moisture may also be a factor as it may take less effort to locate a containment line on a naturally moist place. Ground radar may be used to map the soil moisture and so may be factored in as a variable. Containment lines may still be installed in dry places because water can be brought in the winter and stored. When a fire approaches, the ground may be made wet by the stored water before it arrives.

[00577] A Smart Containment Line Design System is described. The containment line design system may be a computer assisted manual system that may assist people to design effective and cost effective containment lines in specific locations. The containment line design system may use design decisions and collected data with a machine learning algorithm based design system to increasingly automate the containment line design process.

[00578] The containment line design system may contain the following elements (maps, modules, models etc.):

[00579] Asset location map may contain the location, type, importance and value of the assets to be protected. The value of the assets may impact the overall cost of the containment line for it to be cost effective: valuable assets may have more spent on protecting them.

[00580] Fire data module may contain information about fires in the location of the containment line, and include such information as the wind directions of the previous fires and the prevailing wind directions in summer. [00581] 3D topographic model. A 3D mapping system may be used for creating 3D maps for helicopters to create 3D maps for containment lines. These highly accurate maps may provide the input to accurate wind modelling around the containment line.

[00582] 3D fuel load map. The most accurate map data may come from satellites and may hold information about the fuel load in squares across the area that is to be protected.

[00583] Observation (sensing) tower model. The containment line and surrounding areas should preferably be under the direct observation of towers to provide minute by minute fire information. The positions of observation (sensing) towers can be calculated by using the 3D topographic map. The observation (sensing) towers may record everything they can about the fire, the flow of embers, the propagation of the fire along the ground, firefighting activity etc.

Each vehicle and firefighter may be given a locator device that will allow detailed records of the fire and how it is fought to be provided to the system. This information may be used to scientifically update the strategies to defend fire containment lines.

[00584] Vehicular traffic model. Safe egress and ingress should be provided and there should be sufficient space for firefighting. There may need to be 2 lanes or if one lane, regular passing areas and firefighting areas. This model may calculate the number of vehicles and people needed to defend the containment line and make sure that the containment line can cater for those vehicles and people.

[00585] Water logistics module. Siting containment lines near to a water supply will enable that water to be used to fight fires. When local water is not available, large durable water tanks may be needed for many containment lines to provide firefighters with water to fight fires and allowing helicopters to reload fire suppressant and fight fires with minimal transit times. The water tanks need to be sited in places where they can be easily refilled e.g. by rain capture, pumping water from another source and by water tankers which means that there should be suitable road access. Tanks may be refilled outside of the fire season to reduce vehicular access in fire season and provide year-round work. Water may be pumped into these tanks. One design of water tank may be used to lower cost by having significant numbers manufactured. The water tanks need to be located so as not to block vehicular movement and to allow water to be safely and conveniently provided to the firefighters where they need it. Buried gravity fed pipes may be required to provide water to the best firefighting locations.

[00586] Ember modelling module. This model may predict the flow of airborne embers and how they interact with baffles and the topography of a containment line. This module may be improved after every fire as the system may be able to record and track the flow of embers from the observation (sensing) towers. The majority of embers flow approximately parallel to the ground, and mostly at a height of less than 8 metres. The distance they go in the air is affected by wind speed. The distance travelled by embers can be reduced by causing wind turbulence where they can fall to the ground. It is critically important that the embers land in a place where they self extinguish (e.g. on moist ground) and where it is easy to extinguish them (best using an automated vehicle) and if they are not extinguished, then there are unlikely to ignite a fire. For example, installing baffles along the top of a steep and narrow ridge may cause the embers to drop down the other side of the ridge where they could be hard to put out. It would be better to have the baffles erected on a wide ridge so that the embers would fall on flat land where it is easier to extinguish them.

[00587] Baffle selection module. This module may suggest a baffle construction plan in which the location, size and types of baffles to be installed is suggested in a 3D topographic plan which may maximize the protection from embers. Each baffle type may be aeronautical modelled in a number of different topographies to see what the effect is of the baffle on creating turbulence and the dropping of embers. Information from this submodule may be used to suggest the optimal location for baffles along the containment line. However, there may be a conflict between the location of baffles and other requirements, such as water tanks.

[00588] The aesthetic design of the baffles may be important where they are visible. It might be necessary to have different baffle designs and layouts to produce an aesthetically pleasing containment line design.

[00589] Any conflicts may be resolved by machine learning systems to develop increasingly intelligent and automated containment line design processes.

[00590] A firefighting manual may be provided for each containment line. This may be in 2 parts: a general part which is common to all containment lines and a part that is specific to each containment line, e.g. telling someone if they need water where to go. As data is collected, a fire fighting control system may be established that may optimize fire fighting efficiency e.g. by scheduling the refilling of vehicles with water tanks to ensure that there is not a queue to refill vehicular tanks.

[00591] The herein described driverless fire control vehicle is an autonomous fire fighting vehicle that is both a water tank and an autonomous fire fighting vehicle. The vehicle may include large low pressure tyres for putting out embers by crushing them as well as, optionally, spraying the embers with water. This vehicle may be fitted with an implement to disturb the soil. The vehicle may then wet the soil to stop the propagation of the fire along the ground prior to the arrival of the fire. The vehicle may be monitored so that when an obstacle is encountered, a remote fire fighter can take control.

[00592] Described is a bushfire information research database for collecting, formatting and organizing existing GIS, bushfire and relevant weather information into the bushfire research database.

[00593] Further data may be collected from the following.

[00594] Observation (sensing) towers may be set out in a grid to collect further data. The sensing towers may capture data associated with one or more of a) Weather, b) Ground moisture, c) Cloud height, d) Charge in the air, e) Lightning strikes and f) Multiple fire detection sensors using stereoscopic arrays, cameras with different focal lengths, triangulation to maximize location accuracy and minimize false positives and negatives etc.

[00595] The lightning strikes may be tracked using one or more of visual, magnetic and radio sensors. For example, the location of a lightning strike in 3D may be used to measure elevation and groupings of lightning strikes, path from cloud to ground may be determined, polarity of the strike may be determined, and topography may be determined.

[00596] Fire detection sensors may include one or more of visual, infrared, smoke detectors (visual and chemical) and sensors for correction of atmospheric distortion

[00597] Satellites may collect further data to determine/generate one or more of a) ignitions, b) fire maps in real or near real time, c) fuel loads and d) cloud and weather information.

[00598] The smartphone application program may collect further data to enable one or more of a) efficient reporting of fires, b) monitoring of fire related information such as fuel loads, c) location of people in emergencies and d) communications to and from people

[00599] Data may be collected by operations to enable 3D mapping to enable helicopters to operate safely at night. This data may be used to determine ground moisture, e.g. from aerial survey using ground penetrating radar. This data may include lightning strike inspection data to see if ignition occurred. This data may include data measurements made inside clouds. [00600] Operations activities may be stored in the bushfire research database.

[00601] These operational activities may include data collected by operations, such as, for example, 3D mapping to enable helicopters to operate safely at night, ground moisture from aerial survey using ground penetrating radar, lightning strike inspections to see if ignition occurred, measurements made inside clouds

[00602] These operational activities may include lightning reduction activity.

[00603] These operational activities may include inputs to drone design and testing of drones.

[00604] These operational activities may include fire suppression missions, such as those undertaken by helicopters, planes and/or drones.

[00605] These operational activities may include sensing tower construction and maintenance by helicopter.

[00606] Innovative aspects of the emergency response system include one or more of the following: a comprehensive emergency response system; a comprehensive collection of existing data into one integrated database; focus on the rapid detection of fires and the quick extinguishing of these fires by large helicopters delivering fire suppressant rapidly with pinpoint accuracy enabling multiple fires to be extinguished with the one sortie; an App for the public to quickly and efficiently report fires; instrumentation on helicopters to ensure the fire is extinguished; systematic integration of detection, fire response, fire containment and lightning prediction/response; layering of the fire detection system to detect all fires however ignited; systems to minimize false positives and negatives in fire detection; new ways to deliver fire suppressant to fires, such as the smart nozzle; using inspection and attack drones to fly beneath thunderstorms to detect and attack ignitions; smart scheduling of the fire responses; calculation of the cost of not responding to a fire; building of strategically placed water storage in dry areas to enable low transit times from fires to water; developing very accurate 3D maps of the firefighting terrain to allow helicopters to function safely at night; prepositioning of aerial resources based on prediction of dry thunderstorms; releasing of glider/drones from planes; using multiple sensors and analysis to accurately locate and classify lightning strikes; systems to predict the probability of ignitions from particular lightning strikes; satellites to collect information about the structure of thunderclouds; observation (sensing) tower design and the layout of towers and their ability; use of satellites and observation (sensing) towers to deter human lit fires; using a combination of spectroscopic camera arrays and triangulation to pinpoint the location of ignitions or potential ignitions; systems to assist pilots to deliver fire suppressant with pinpoint accuracy to small fires; fire suppressant devices; effective containment line design software to enable the construction of effective containment lines; actions to stop or reduce the frequency and intensity of lightning strikes.

[00607] According to one example described herein, there is provided a system that can prevent fires, reduce intensity of fires, and/or predict the location of ignitions. The system may detect fires within minutes of ignition (the most damaging fires are ignited by lightning, often in remote areas). The system may quickly extinguish ignitions while still small. The system may be used as part of an effective national fire containment line strategy.

[00608] The system described enables an integrated, comprehensive, pre-emptive plan to become possible. The system described detects new ignitions within minutes of ignition using a combination of satellites, observation (sensing) towers, drones and aircraft combined with advanced analytics software to track lightning strikes and detect all fires however ignited. The system described may utilize a necessary number of suitable helicopters that can fight fires at night, and which may be positioned close to predicted dry lightning storms to respond quickly and decisively to new ignitions. The system may utilise an installation of a network of large durable water tanks in places with little available water to increase the effectiveness of the helicopters by reducing transit times to refill with fire suppressant. The system may utilise effective and cost-effective fire containment lines to protect infrastructure, assets and property, and reassure the public, in the unlikely event that a fire is not extinguished quickly.

Industrial Applicability

[00609] The arrangements described are applicable to emergency response industries and particularly for the emergency response systems and devices industries.

[00610] The foregoing describes only some embodiments of the present invention, and modifications and/or changes can be made thereto without departing from the scope and spirit of the invention, the embodiments being illustrative and not restrictive.

[00611] In the context of this specification, the word “comprising” means “including principally but not necessarily solely” or “having” or “including”, and not “consisting only of”. Variations of the word "comprising", such as “comprise” and “comprises” have correspondingly varied meanings.