Field of invention

[01] This invention relates to bushfire path prediction and in particular to a method of predicting the likely path that a fire will follow towards a potentially affected area.

Background to the invention

[02] A significant threat to real estate in hot and dry climates is posed by bushfires.

Often, the threat is exacerbated by the lack of an integrated early warning system to those under threat.

[03] United States patent number 7,834,771 to Lee discloses a fire detection system making use of an infra-red camera to capture an image and a analysis software to determine the likelihood that the image is one of a fire. Warnings are outputted to disaster prevention authorities. The system includes a laser range finder for helping establish the exact location where the fire has broken out. By making use of a plurality of detectors, an average point of location of a fire breakout can be calculated.

[04] It is found that as a bushfire grows in size, it affects localised airflow patterns and sets up fluctuating air currents that are difficult to monitor. The prior art does not appear to address a way of using real time data of actual wind and atmospheric conditions at large in a relative microclimate in order to predict the speed and direction of a moving fire front.

Objects of the invention

[05] It is an object of this invention to address shortcomings in the prior art. In doing so, it is sought to provide a method of predicting the path that a bushfire is likely to follow.

[06] It is also an object to provide means for issuing an early warning of bushfire danger and the predicted path of a fire front particularly in areas having low population density, where there may be no one present in the immediate facility to send out an immediate alarm.

[07] Further, an object is to provide means and methods for facilitating continuous communication between occupants or residents of an affected area and fire-fighting services.

[08] Another object is to provide a method of predicting the speed of approach of a fire front towards a threatened location taking into account localised winds. [09] The preceding discussion of the background to the invention is intended to facilitate an understanding of the present invention. However, it should be appreciated that the discussion is not an acknowledgement or admission that any of the material referred to was part of the common general knowledge in Australia or elsewhere as at the priority date of the present application.

[010] Further, and unless the context clearly requires otherwise, throughout the description and the claims, the words 'comprise', 'comprising', and the like are to be construed in an inclusive sense - that is to say, in the sense of 'including, but not being limited to' - as opposed to an exclusive or exhaustive sense - that is to say meaning 'including this and nothing else'.

Summary of invention

[011] This invention provides for detectors mounted at various known locations on or around a property to be in communication with a computer processor module and in turn with a mobile device loaded with a software application program ("app") which is operable to communicate with residents in an area having significant bushfire risk, enabling them to be alerted and prepared year round whether present or absent from their property. The detectors are selected to feed micro-climatic data automatically or on demand to a designated recipient for analysis and response, and to compare incoming data with reference criteria determinant of the level of bushfire risk or an actual bushfire event. The combination of these devices facilitates continuous communication between residents and their fire management services.

[012] Hence, according to a first aspect of the invention, there is provided a bushfire detection and behaviour prediction method comprising the steps of a. detecting conflagration using a combustion detection device; b. measuring selected atmospheric condition values at a first measuring station at a known location remote from the conflagration; and c. from the measured values, calculating a predicted path of travel and speed of a fire front associated with the conflagration.

[013] In a preferred form of the invention, the atmospheric conditions are wind speed and wind direction. The method preferably also includes measuring air temperature at said known location. Further preferably the method includes measuring humidity of the air at said known location.

[014] In a preferred form of the invention, the method includes analysing data from the combustion detection device and determining whether it meets criteria representative of a bushfire origin.

[015] An example of a combustion detection device that is utilised in the method of this invention is a thermal imaging device.

[016] In a further preferred form of the invention the step of detecting conflagration comprises operating a flame detector device or capturing thermal images of a landscape. The thermal images may be captured by operating a detector sensitive to radiation in the near infrared range of the electromagnetic spectrum.

[017] The method preferably further includes the step of communicating data relating to said images to a data-receiving device located at a control station.

[018] In a still further preferred form of the invention, the method includes providing a plurality of measuring stations, measuring atmospheric condition values at stations located in a pre-defined geographical proximity to a first station at which conflagration is detected and communicating said values to the data-receiving device.

[019] In an embodiment, the step of communicating the measured values or data comprises uploading the data via a global network of computers, for example the "internet".

[020] In a further embodiment, the method of the invention comprises establishing a virtual private network of user devices. The method may further include the step of communicating the measured values or data via the virtual private network.

[021] In an embodiment of the invention, the method includes transmitting and receiving data over a wireless communications network. A preferred network is an AM radio network. In another embodiment, the network is one that operates in the 27GHz bandwidth.

[022] According to a second aspect of the invention, there is provided a hyperlocal detection network comprising a plurality of local networks each associated with a designated geographic locality, communication means between the local networks enabling data exchange between them, wherein the data comprises appearance data pertaining to a locality and atmospheric condition values from a station proximate the locality.

[023] According to an third aspect of the invention, there is provided a bushfire path predicting method, comprising the steps of selecting a location to monitor, operating a computer configured to collect, at timed intervals, real time data associated with the location, said data including appearance data of the location, and data relating to microclimatic atmospheric conditions current at or proximate to said location; and causing the computer to run a program to identify, from said appearance data, whether characteristics of a bushfire are evident; and where evident, predicting, from said second data, a path of travel of a bushfire associated with said appearance data.

[024] In a preferred form of the method, the computer collects the real time data from first and second monitoring stations proximate the location and located to be spaced from each other.

[025] Further, according to the invention the step of identifying evident bushfire characteristics preferably includes causing the computer to compare the most recently collected with data collected from at least one previous interval.

[026] The method preferably also includes computing, on the basis of the real time data, a bushfire danger rating for an area proximate said location.

[027] In a preferred embodiment, the method comprises the step of estimating, from at least the appearance data, a bushfire front length.

[028] In a further preferred embodiment, the method includes predicting speed of travel of the fire front.

[029] The method preferably also includes the step of predicting at least one other locality likely to be in the path of said fire front.

[030] In a further preferred from of the invention, the method includes programming the computer to issue an alert if said characteristics of a bushfire are evident.

[031] Further, according to the invention, the computer preferably is programmed to select a recipient device to which it issues said alert, intended for a known designated user of said device. [032] The computer preferably is programmed for selecting the recipient device on the basis of the path prediction.

[033] In a still further preferred from of the invention, the method includes establishing communication between the computer and the recipient device, using a global network of computers.

[034] Alternatively or in addition, according to the invention, the method includes establishing communication between the computer and the recipient using a wireless communications network.

[035] In a yet further preferred from of the invention, the data relating to atmospheric conditions comprises wind speed and wind direction data.

[036] In an embodiment, the appearance data relates to a landscape at the location being monitored, as viewed from a monitoring station.

[037] In a further embodiment, the appearance data comprises measurements of incident electromagnetic radiation. The method preferably includes processing said radiation measurements for generating from them thermal image data pertaining to the landscape.

[038] The method of the invention preferably further includes collecting data relating to a plurality of locations at substantially the same time.

[039] Preferably, the method also includes collating data relating to landscapes located proximate the location at which bushfire characteristics were first identified.

[040] The method preferably further comprises communicating the data to a first repository and, from the data, using data processing means for computing a fire danger rating for each of a plurality of locations.

[041] According to the invention, in a preferred embodiment the method includes the step of using the fire danger ratings for defining a bushfire danger zone.

[042] The invention includes the step of communicating predictive data relating to the fire front and path to any one or more of recipients devices in a set comprising devices associated with persons resident or known to work in the danger zone, their nominated relatives, owners or custodians of assets in the danger zone, emergency services having jurisdiction over or responsibility for territory in the danger zone, and devices detected to be within the danger zone.

[043] A fourth aspect of the invention provides for a wildfire monitoring station, comprising real time data-capturing equipment comprising appearance data-capturing apparatus configured for capturing appearance data of a landscape observable from the station, atmospheric data capturing apparatus, and means for communicating in real time said data or data derived therefrom to a computer processor.

[044] The monitoring station preferably comprises a housing and mounting means for operatively mounting to said housing the real time data-capturing equipment.

[045] In a preferred embodiment, the appearance data-capturing equipment is pivotally mounted to the housing.

[046] Further preferably, the appearance-data capturing apparatus comprises at least one measuring device for measuring incident electromagnetic radiation. The measuring device may include thermal image-generating means configurable for generating a thermal image of the landscape from the incident radiation.

[047] At the monitoring station the atmospheric data-capturing apparatus preferably comprises wind speed measuring means and wind direction determining means.

[048] In an embodiment of the invention, the monitoring station comprises an air temperature sensor. The monitoring station may further comprise an air humidity sensor.

[049] In a preferred form of the invention, the monitoring station comprises location- determining means.

[050] According to another embodiment, the monitoring station has the processor housed in the housing.

[051] In an alternative embodiment, the processor is located to be remote from the station.

[052] In a still further preferred form of the invention, the computer processor is adapted for identifying from said appearance data, characteristics of a bushfire on the landscape. [053] The processor is preferably programmed to compare the thermal image data with a threshold standard indicative of a bushfire and output a notification should the threshold be surpassed.

[054] Still further according to the invention, it is preferred that the monitoring station comprises wireless communication means configured to send data to remote receiving means. The data preferably includes identification information and station location co-ordinates.

[055] The wireless communication means may be configured for communication on a network operating at 27GHz.

[056] In a further aspect of the invention, a bushfire monitoring system comprises a local area network and a wide area network of computational devices, the local area network comprising a plurality of real time data collection stations and communication means configured to be in data communication with said collection stations and with devices on the wide area network, characterised in that the data collection stations comprise appearance data- capturing means adapted to capture appearance data of a selected landscape, and atmospheric condition capturing means configured for capturing atmospheric condition data for association with the selected landscape.

[057] The bushfire monitoring system preferably comprises data processing means configured to receive captured data and programmed to identify, from data received, characteristics of a bushfire, and communicate notification thereof to a device associated with the wide area network.

[058] The processor is desirably instructed to calculate a probability that the characteristics are those of a bushfire and compare the probability with a threshold value representative of the presence of a bushfire event.

[059] According to the monitoring system, the processor is preferably instructed to calculate a hazard rating from the characteristics according to a predetermined standard rating scale.

[060] In a preferred form of the invention , the processor is programmed and configured for outputting an activation alert to a recipient device for activating a predetermined bushfire response plan. [061] Still further, the monitoring system of the invention is configured so that the recipient device preferably displays or sounds instructions to a user thereof according to the response plan.

Brief description of drawings

[062] In order that the invention may be readily understood, and put into practical effect, reference will now be made to the accompanying figures. Thus:

Figure 1 shows in diagrammatic view a preferred embodiment of the system of fire detection of the invention.

Figure 2 is a diagram of a high level operational view of the system of figure 1.

Figure 3 shows views of a preferred embodiment of the sensor station of this invention: (a) front, (b) top, (c) side and (d) cross section of (c) along the line of symmetry of figure 1.

Detailed description of an embodiment of the invention

[063] The bushfire warning system of this invention links individual, location-based user- managed local area networks to create a wide, hyperlocal detection network, providing simultaneous wireless communication and pairing between all data collection modules within the network. The wireless connections may be established through standard Wi-Fi and 4G mobile telephone connectivity, or any other suitable wireless protocol or frequency ranges. Where not practical to use mobile telephone communications, the connections may operate on radio frequencies such as in the 27GHz band or amplitude modulation (AM).

[064] The invention enables live, accurate and fast delivery of information to those affected during a bushfire emergency. By setting up monitoring stations having detectors and data collection modules ideally on every property within a demarcated management zone and configuring these for constantly communicating with each other, the system vastly increases the area of automatic detection of a bushfire in its early stages. It also facilitates the prediction of a fire front path and its rate of movement, thereby enabling rapid response by both residents and emergency services, whether this is to evacuate early or stay and fight the fire according to predetermined emergency management plans. [065] The system of the invention thus comprises a plurality of local (personal or private) detection networks linked with each other and to a local authority module to form a zonal macro wide area network.

[066] Each local network comprises a data-capturing station in wired or wireless communication with a home module. The data-capturing station is equipped with either or preferably both a landscape appearance data-gathering device, such as a thermal imaging camera unit, and atmospheric conditions monitoring apparatus, including in particular wind speed measuring and direction sensing devices. Wireless communication may be via a suitable wi-fi standard such as 802.11g/n/ac, or via UHF or other suitable radio frequency channels, as permitted by law and subject to distance and topographical factors. Commercial mobile telecommunications networks, preferably to the 4G standard, as well as networks utilising other communications standards and protocols, may also be used where appropriate, available and reliable.

[067] The appearance data-capturing unit is connected to be in communication with and to feed image data captured from its local field of view to a data processor programmed to identify, from the image data, the existence of a thermal hot spot that has characteristics associated with a bushfire. The processor compares data from successive and historical images and is programmed to respond to a reference difference between the images, preferably supported by an absolute criterion such as a threshold temperature of a detected hot spot, for example 240 degrees Celsius. The image data typically comprises measurements of incident electromagnetic radiation emanating from a flame, as detected by one of more of the following devices or combinations thereof: ultraviolet radiation detectors, near infrared (NI ) array or visual flame detectors, infrared (IR) detectors using thermographic cameras, such as thermal imaging cameras; UV/IR detectors, IR/IR detectors; multi-frequency flame detectors, such as IR3 detectors; and visual and video such as CCTV and webcams.

[068] The processor may be locally installed as a component of the data collecting station, or it may be located remotely at a home module, or even at a regional control module, depending on the system architecture determined to be appropriate to the location and community served by the system.

[069] The atmospheric conditions monitoring equipment detects not only wind speed and direction, but also air temperature and humidity data. Data feeds to the processor are programmed to take place at a selectable time interval. Should the appearance data be determined as characteristic of a possible conflagration, the processor is programmable to increase its scanning frequency of the wind sensing device (so that the interval between data- capture events decreases). In a preferred embodiment, the wind sensing device is ultrasonic.

[070] In the case where the appearance data is collected by a thermal imaging camera unit, the unit may comprise more than a single camera, since two or more thermal imaging cameras are positionable to cover a wider angle of view than one. However, this choice can depend on the locality and its topography. Efficiencies may be increased by gathering and comparing data of the locality being monitored when captured from units located on neighbouring properties, instead of duplicating units. However, this choice would be subject to redundancy factors and ease of communication between local networks on adjoining properties and between each local network and its zonal network. It is desirable that at least some of the stations monitor landscape portions being monitored by at least one other station, so that a degree of overlap occurs. This enables corroboration of data and increased accuracy of predictions, when atmospheric data from stations set up to monitor an overlapping landscape can be taken into account.

[071] Preferably, motor-driven, automatically-pivoting cameras are employed for sweeping as wide an arc as possible. The configuration choice at each locality will depend on energy requirements, source efficiencies and cost factors at large. Solar power and battery packs are preferred to be deployed, when suitable conditions pertain. Alarm signals are generated when communication is lost, or when battery replenishment is not occurring and charge is getting low.

[072] In addition to the appearance data unit and atmospheric conditions sensing unit, the data-collecting station may include a smoke detector or particle detector device. These inclusions are advantageous in undulating localities, where electromagnetic rations, such as from thermal images (for example) may not be received until a fire crests a ridge (should a thermal imaging unit not be present or operational on the ridge).

[073] The home module will do calculations regarding wind speed and direction to calculate the bushfire danger rating, the risk of the fire detected and the rate at which it might hit the next property. All the information gathered (wind speed and direction, humidity and thermal detection) will be sent over the wide, hyperlocal network to the fire services, for them to act on. The home module may be a standalone unit having a housing in which are installed a PC board, and communications and processing hardware and operating software. Alternatively, the home module may be provided in the form of an application for running on a home computer located at the premises being protected having suitable wireless or Ethernet connections with the detector module.

[074] Data captured, as well as calculations and communications performed by the home module are backed up to the regional or zone module at periodic intervals, or are backed up to Cloud-based storage. This ensures that data collected/captured by a local module is not lost, even if the home module is destroyed by fire. This enables the review of data in post-fire investigations.

[075] The software for managing the system comprises a mobile telephone application ("app"). The app provides a remote connection to the local network. It allows a user to be notified of an emergency whether absent from or present at their property. It has 7 main features that allow a user to prepare, act and stay connected before during and after an emergency. a. Emergency alert -

This can come in the form of a notification, such as by text message, of a very high to catastrophic bushfire danger rating being broadcast by a fire management authority and / or when a real fire is present and has been detected by the thermal camera of the data collection module on a local network. b. Live updates before, during and after an emergency - Live updates may be and preferably are provided by both the personal network (wind speed and direction, humidity levels, temperatures, flame detection) and the fire management authority. During the year and during the bushfire season fire management authorities send notifications to residents to help them prepare for the fire season. By way of example, these may include "Clean leaves from gutters, roofs and downpipes", "Remove excess ground fuels and other combustible material", "Have you got your bushfire survival kit?". Live updates during an emergency may include announcements such as "Your closest safe zone is 2km away at Smith Park, 123 Smith Rd, Smithville", "A fire is approaching your property at a fast rate. Please commence implementing your survival plan"; "It is advised that you evacuate immediately if you have a safe route to do so". Live updates after an emergency may include "A fire management authority officer will be reaching you in the next 24 - 48 hours to discuss your next steps, damage and loss to you, your family and / or property". c. Survival plan -

The user draws up a survival plan in the bushfire off-season and this is preprogrammed into the system, with hard copies being retained as appropriate for use in the event of data or equipment loss. The plan documents a resident's action steps to be taken whether he has chosen to prepare to evacuate, or to stay and protect their property if fire threatens. The plan is loaded and activated before or during an emergency, to remind the user of action steps required and help them make confident decisions for their safety, health, wellbeing and survival. The user may choose to have neighbours, friends and family on the network notified should their survival plan be activated. The designated proposed recipient may choose to receive or decline such notification on their nominated communications device. d. Test

Users test their survival plan and any equipment or installations, such as existing smoke detectors, roller shutters and sprinkler systems that they have synchronised ('synced") to the network via the app and their home module. e. Sync

Users sync existing products such as existing smoke detectors, roller shutters and sprinkler systems to the home module via the app. f. Friends and family -

Users stay connected to their friends and family on the network and notify or are notified if they have activated their survival plan or an emergency is present. g. House Lockdown -

House lockdown activates all the equipment and installations that have been synced to the network. This can be done via the home module or remotely via the app from any location worldwide (provided communications are available).

[076] An exemplary embodiment of the system of the invention will now be described with reference to figure 1, where it is depicted for convenience as comprising three levels, 1, 2 and 3. Central to the invention is level 2, comprising a plurality of local area or personal networks 102 (marked PN). These PNs are interconnectable (104) by means of a global network of computer devices, such as the internet. In other embodiments, alternative forms of network communication may be employed. These interconnected local networks of level 2 define a hyper-local detection network (HLDN) 118. The HLDN is subscriber-based, not permitting general unrestricted access to disinterested parties on the internet. The HLDN also interfaces with one of more networks operated by emergency management services 116. A separate level 3 network includes the emergency management services with local government or state government organs and service providers, as the case requires and is determined by the relevant authorities from time to time and for a particular emergency event.

[077] Working from the ground up (as it were), level 1 illustrates the components of a typical local area network or personal network (PN). The network has a number of fire detector stations 106 connected to a home module 108 and in communication with an application 110 running on a personal mobile device such as a tablet computer or smartphone 112. Home module 108 is an internet-capable computer device, for example a personal notebook or desktop computer having at least 2GB RAM and a 250GB hard drive, preferably running an operating system such as Windows 7 or a later release. Suitable alternative operating systems may be advantageously employed in other embodiments or in other individual PNs or devices in the structure of level 2. Older operating systems such as Windows XP are not excluded, but are less preferable nowadays. The main requirement for the computer module is for it to be internet capable for receiving and communicating data received from devices on level 1 to local networks on level 2 (including systems that may also be on level 3).

[078] A detector station 106 has devices and instruments that will be described in detail below with reference to figure 2. It functions to capture data relating to environmental conditions associated with a selected landscape and to make the data available for downstream processing. The processing results in a determination whether there is a fire in the offing and an assessment of the hazard it poses to person and property. Individuals and authorities receiving the assessment are then put in a position to make decisions regarding their own safety and to implement escape plans or plans to fight the detected fire.

[079] Level 3 represents a layer of emergency management systems operated by local or regional authorities, such as fire-fighting services and civil defence. The PNs relating to property within the jurisdiction of a particular local authority desirably connect to the emergency or fire- fighting services of that authority. The local authority is in turn connected to neighbouring local authorities and to a regional authority, or appropriate wider authority. For convenience these authorities are generally depicted as a single mass at 116.

[080] The operation of the system will be described with reference to figure 2.

[081] Beginning at level 1, detector station 106 from figure 1 includes, in this example, the following instruments and detection or measuring devices: a. A thermal imaging camera 124 (alternatively an infra-red camera) b. A wind direction and speed sensor 126, such as an ultrasonic transducer assembly of a kind known in the art; for example, the MU ATA MA40S4S SENSOR available from elementl4 Pty Ltd., of Angel Place Level 8, 123 Pitt Street, Sydney, NSW 2000, Australia c. A humidity meter 130. d. A temperature sensor 132 that senses the local air temperature. A suitable example is a combined humidity temperature sensor module, known as the DHT22, available from Freetronics Pty Ltd, PO Box 7067, Croydon South, Victoria 3136, Australia.

[082] The station also includes a GPS receiver 134 by means of which its position is determined and included in the data collected and transferred to the home module 108. This ensures that the location of the station that is reporting the data is accurately and automatically logged by downstream processing units. This reduces the risk of erroneous locality data being collected in the event of the station being relocated for any reason, for example by the property owner.

[083] The thermal imaging camera 124 periodically communicates data of the landscape on which it is trained. The interval of transfer may be selected by the user. The default is 3 minutes (20 times per hour). This interval length provides sufficient time for significant differences to be noted between successive appearance data sets, but is not too long for a blaze to develop uncontrollably before it is detected. For overall system efficiency, the local authority 116 may determine and direct a particular minimum frequency at which data must be reported and analysed in order to ensure early detection of fire hazards. This frequency may also be established by legislation. [084] A software algorithm 128 in the detector module 106 analyses the data and looks for characteristics indicating the start of a conflagration 142, in which event it collects environmental data 144 by scanning outputs emitted by each of the connected devices at the detector station. The data is processed to predict the direction of travel of the fire front and its speed and intensity. The information is communicated via 4G, or Wi-Fi, F or other wireless communication signals to home module 108, which, at step 146, in turn communicates it to neighbouring properties connected to form part of the hyper-local detection network and at step 148 to the emergency or fire-fighting service 116 having jurisdiction. The emergency management authority systems collate the data received from home module 108 and the other PN modules in the network to build a less localized, higher level view of the conflagration and its predicted path and speed. Factors such as fuel availability values are called up from data storage services and these are used to further improve the accuracy of predictions.

[085] The home module comprises a personal computer, laptop or notebook 108 that has been programmed to raise the alarm locally by messaging the mobile device/s of the immediate local resident/s and residents on neighbouring properties. On dispatching data representing a suspected conflagration, as described above, the computer performs a risk analysis on the data using locally loaded software (step 150) and issues its findings in an on-screen notification to any user present, and broadcasts the findings to other computers on the same network and to mobile devices associated with the principal user and members of the user's homestead. It also transmits notification of its assessment to other interested parties named by the user, and automatically to near neighbours and designated disaster management authorities.

[086] The home module computer is also programmed to activate the pre-programmed emergency response and survival plan 152 of the homestead where it is located. Activation steps are selectable by the household concerned, unless automatically programmed, for example in the absence of a user command. A menu of recommended actions is provided to the user via the programming. Typically activation includes texting the householder and residents individually and outputting a copy of the steps of the emergency response plan. The steps are set out interactively, requiring human input to acknowledge and confirm that a step has been implemented.

[087] The computer of the module is also programmed to sound an audible alarm 154, accompanied by visible alerts such as flashing lights. The computer is further programmed to put the buildings where it is housed, or other assets of the landowner, into a lockdown mode 156. The computer is thus operatively connected to be in communication with protective equipment such as fireproof roller blinds, fire-fighter water pumps and spray controllers and to activate motors or mechanisms operating these, at a predetermined time. The implementation time is calculated depending on the speed of approach of the fire front, and whether lockdown mode has not yet been activated by human intervention. In an example, in the absence of human intervention within 6 minutes of the alarm being sounded, lockdown mode is implemented automatically.

[088] When a user at a threatened homestead activates their survival plan, a notification is generated and transmitted to that user's selected and pre-listed group of recipients, such as friends, relatives and next of kin. The stem will send news, this time automatically, to neighbours in the path of the fire front in order to avoid duplication of efforts and efficiency of communication at times when infrastructural limits may be under severe pressure.

[089] In parallel to the communications passing between the home module and devices on the emergency management computer network 116 and near neighbours having their own PNs, the process features described above are echoed and replicated on the or each mobile device loaded with the system software mobile app. This helps ensure that those absent from the immediate locality of the approaching fire are notified and kept informed. It also keeps those persons present informed without the need for them to be stationed at a PC or to carry an inconvenient laptop or notebook device in the circumstances. This means the user of a mobile device 170, loaded with the system software mobile app 172 is potentially able to receive the earliest alert signal 174 and risk assessment 150, and be prompted to activate their survival plan 152'. Basing a decision on the risk assessment, the user of the mobile device may decide remotely on activating house lockdown 156', while also receiving live updates 158' and maintaining connectivity with their circle of chosen associates and automatically designated neighbours and emergency management services (see 160').

[090] A non-limiting example of a data-collecting station for use in the system will now be described with reference to figure 3. Here the station is in the form of a module (marked 20) illustrated in front view in figure 3(a). It comprises an upper hollow body 22 having a top outer housing 1 made of aluminium and a top inner housing 2, also of aluminium, and a rubber seal 3 separating but forming a seal between them. The body is symmetric with the same arrangement of outer and inner housings 4, 5 to the left of its vertical axis of symmetry, which is not shown, but coincides with line 30 when extended in both directions. The housing is shown from above in figure 3 (b) with like parts carrying like numbering. It defines four hollow protrusions 32, 34, 36, 38 extending from a central hollow space 40 between them. The protrustions house respective wind transducers, to be further described below.

[091] Detachably connected to the upper hollow body via a neck 24 is a lower housing 26 of aluminium. Centrally located is an aperture 28 through which a thermal camera lens is mounted. The lens 9 is sealingly engaged in position by means of a silicone sealing ring 6. Being a wide angle lens with a prominent convexity giving a 170 ^° field of vision, it protrudes partially, as seen at 9 in Figure 3(c) (side view). Lens 9 is part of the thermal camera 11, which is shown in cross-sectional side view in Figure 3(d), mounted in the housing by means of a silicone inlay.

[092] Lower housing 26 in turn is connected to a gantry arm 42 via a pivotal joint assembly 46, allowing the user to select the direction of orientation of the lens above. The gantry arm is mountable to a supporting structure using flange 44 and threaded connector 48. The supporting structure may, for example, be a post or a wall.

[093] The four protrusions 32,34,36,38 in the top aluminium housing 1,2 are shaped to allow wind to pass through whilst ultrasonic waves are being bounced between the transducers mounted to them. These transducers 7, 8, 10, 12 are oriented to detect temperature and monitor wind speed and direction. This information is calculated and sent to the home module for further distribution of information to fire services/local authorities and the mobile app operated on the user's smartphone.

[094] With reference to figure 3(d), there are shown the positions of opposed transducer pair 10 and 12 and a humidity sensor 13 in the upper housing portion 22. The humidity sensor measures moisture conditions in the monitored location and the data is used in calculating the likelihood of a bushfire occurring. This calculation will help aid fire services with fire danger ratings.

[095] Also housed within upper housing portion 22 is a printed circuit (PC board) on which are located the circuitry components for managing data collection and communication from the housing to the data-receiving module. The components include the following operatively connected items: a. A GPS module for confirming the position at which the data is being collected.

Including this feature enables the station to be relocated by the operator or the local authority should this be considered desirable, for example in the event of trees or new building structures obscuring the field of vision; b. A USB port for use in uploading software updates or manually downloading data, or performing fault diagnostics or other system testing; c. A SIM card for enabling mobile communications over a digital subscriber network; d. A K-band transmitter configured at 27GHz for communications of the data collected; e. A humidity and temperature measuring module as previously described.

[096] Lower housing portion 26 is cut away to reveal an inner silicone housing layer 15 for protecting internal components against the weather. These components include the thermal imaging camera 11 and associated wiring and communications hardware.

[097] Housed within hollow gantry arm 42 is a 12V solar-rechargeable lithium ion battery, protected by a cover plate 50 in figure 3(c). This helps to ensure the monitoring station stays powered during an emergency if mains power to the installation is lost or not provided, for example for reasons of remoteness. This storage device is connected to existing solar panels or one mounted close to the monitoring system itself.

[098] The PC board may include the home module processor, which is the central brain and distribution device within each personal network. This situation may apply in the case of a user who has no other computer means at his disposal. In this case, the data from other monitoring stations in the network may be communicated to the station housing the processor for interpretation and onward transmission.

[099] The main purpose of the processor, wherever located, is to receive and transmit data from the monitoring stations in its personal network and to interpret the information for the fire services or local authorities and for use in the mobile app of the intended recipient/s. The data communicate may include raw data such as the accurate GPS location co-ordinates of the fire, the size of the detected fire, the ambient air temperature, humidity, wind speed and wind direction. When so programmed, the processor interprets this data to provide a prediction of the rate of progress of the fire front and its likely path. However, the processor of a third level organisation, receiving data from a plurality of personal networks in the vicinity of the detected fire location is programmed to review the local network processor prediction and validate the prediction using data collected and collated from other monitoring stations in neighbouring personal networks by data interrogation of the monitoring stations in these neighbouring networks.

[0100] The processor preferably actuates a local on-board or inbuilt audible alarm that sounds when a fire event is detected. An audible warning alert is sounded to notify recipients when fire ratings are extreme or considered potentially catastrophic in relation to official fire danger ratings. These alarms are operable to alert residents when they are in danger and need to react to an emergency or imminent or potential emergency event.

[0101] The mobile app allows personal network users to be alerted of an emergency whether present or absent from their property. The app acts as a mobile extension of the home module. The interface is designed to educate and keep consumers prepared and informed of the conditions at their property all year round as well as during an emergency. The background colour of the interface is designed to change colours according to the official fire danger rating of that day. This in turn educates and familiarizes personal network users with the ratings and their meanings, as well as what action to take for each.

[0102] These embodiments merely illustrate particular examples of the method, kit and apparatus of the invention providing for early fire detection and path prediction. With the insight gained from this disclosure, the person skilled in the art is well placed to discern further embodiments by means of which to put the claimed invention into practice.