Technical field

The invention relates to a method of navigating an automated guided vehicle for internal logistics in which the current position and inclination of the vehicle provided with a controlled drive and steering connected to an on-board control system are determined, whereupon this detected current position is compared with the desired position of the vehicle on a given reference trajectory of the vehicle movement, and from the difference between the two positions, correction of the vehicle movement is determined, which is then used to control the direction and speed of the vehicle movement.

The invention also relates to a system of navigating an automated guided vehicle for internal logistics which comprises at least one automated guided vehicle provided with a controllable movement system, which is coupled to an on-board control system, wherein the vehicle is further provided with a sensor node for detecting the current position and rotation of the vehicle, the on-board control system being coupled to the sensor node.

Background art

Automated guided vehicles, also known by the abbreviation AGV, are nowadays quite commonly used for internal logistics in industrial complexes and warehouse facilities. To enable these vehicles to fulfil their tasks, it is necessary to determine the paths along which the vehicles are to move, while at the same time ensuring that collisions of the vehicles do not occur where the paths of their movement intersect or where other devices, persons, etc., are located.

Specific physical devices placed in the space where the vehicles are to move are most often used to mark the paths for the defined movement of the automated guided vehicle. Typically, these devices are placed in the floor, on the floor, or in the

SUBSTITUTE SHEET (RULE 26) immediate vicinity of the vehicle path. Known systems and devices include, in particular, inductive, optical and magnetic guide tracks located in the space of the automated movement of the vehicles. However, the fact that the required vehicle path must be marked using physical devices has a major disadvantage, namely that the actual marking of the required path of the vehicles is a costly and time-consuming operation, and, moreover, any change of the required path of the vehicles brings additional costs and time requirements for its realization. Another disadvantage of the systems using physically existing devices for guiding automated guided vehicles is the risk of damage to the known guiding means.

Also known are methods using the so-called virtual guides, which include especially methods of laser navigation of automated guided vehicles. However, even in this case, it is not possible to do without additional measures in the form of the installation of physical devices and elements on the required path of the vehicle. Depending on the specific implementation of this method of guiding automated guided vehicles, systems using virtual guides may use, for example, reflective foils placed at a sufficient height so that the laser beams pass over the heads of the people in the hall where the vehicles are moving, or a set of reference reflective points suitably placed along the path or on the ceiling of the hall, etc.

Ultra-wideband (UWB) or WiFi-based systems have a similar problem, because the installation of large infrastructure, such as beacons, antennas, etc., is necessary for their operation.

The category of systems using virtual guides also includes the use of global satellite navigation systems, especially GPS, GLONASS, BAIDOU and others. However, the use of these navigation systems is in principle limited to the external environment, and even in it only where no major inaccuracies of these systems occur, such as in narrow passages between tall buildings, etc., as is known, for example, from car navigation systems.

Although known methods and systems allow automated guided vehicles to follow specified paths, the common disadvantage of these systems is that superior systems generally do not have up-to-date information about the current position of the automated guided vehicles without the use of additional communication, such as automatic position reporting, etc.

The objective of the invention is to eliminate or at least minimize the disadvantages of the background art.

Principle of the invention

The objective of the invention is achieved by a method of navigating an automated guided vehicle for internal logistics, whose principle consists in that the current position and inclination of the vehicle is detected by a sensor node with an inertial measurement unit and a set of optical sensors, whereby a time stamp is assigned to the detected current position and inclination of the vehicle, which describes the time of data acquisition to determine the respective values of the current position and current rotation of the sensor node, whereupon these data are transmitted to an on-board control system for predictive control of the vehicle movement according to a given reference trajectory of movement of the vehicle and the detected current position of the vehicle.

The principle of the system of navigating an automated guided vehicle for internal logistics consists in that the sensor node is provided with an inertial measurement unit and a set of optical sensors adapted to detect the current position and rotation of the vehicle, whereby the sensor node is further provided with a time unit for issuing a time stamp, which describes the time of acquisition of the data used to determine the respective values of the current position and the current rotation of the vehicle, and the on-board control system is provided with an algorithm for predictive control of movement of the vehicle according to a specified reference trajectory of the vehicle movement and the detected current position of the vehicle.

The invention makes it possible to control the movement of automated guided vehicles in the interior and exterior of industrial and logistics complexes without the need for physically implemented guides or physically implemented guidance markers or even without a satellite positioning system, etc. The proposed solution to automatic guidance of vehicles thus does not require any other technical means in addition to those located directly on the vehicle, out of communication with the control system. The solution is therefore very flexible both in terms of putting the system into operation and in terms of possible continuous or later changes in the vehicle paths, etc. If necessary, the present solution also allows continuous remote monitoring of the current position of the vehicle without additional means. The field of application of the invention is focused on the control and monitoring of movement of automated guided vehicles in the interior and exterior of industrial, warehouse and logistics complexes.

For example, a Microsoft HoloLens (MH) device can be advantageously used as a sensor node. The sensor node is connected to an advanced predictive algorithm for vehicle control based on data obtained from the sensor node. The sensor node, e.g., MH, includes an inertial measurement unit and a set of optical sensors which monitor physical parameters in the vicinity of the sensor node in appropriate directions. For example, the sensor node comprises four video-cameras facing both sides of the vehicle and further comprises one forward-facing infrared depth camera and one forward-facing infrared laser projector. The data from the sensors of the sensor node are correlated and the result of this correlation is an output data block containing information about the current position (x, y, z) of the sensor node and about the rotation (a, b, c) of the sensor node, i.e., angles of inclination, tilting and rotation. This information can be obtained with the use of the sensor node both in the interior and in the exterior, independently of physically existing guides, or independently of the satellite or other positioning system. Since the algorithm by which the position data of the vehicle are determined by its sensor node is complicated, the determination time of these data is variable. It is, therefore, advantageous if a time stamp is added to the position data so that the position data can be used in further steps of the control of the individual vehicles or group of vehicles, etc., where these data are used for two purposes.

The first purpose is the use for predictive control algorithm of the vehicle implemented directly in the control unit of the automated guided vehicle, where, based on the obtained data, the control unit of the vehicle changes the angle of rotation of the steerable wheels of the vehicle and possibly also the speed of the vehicle so that the vehicle follows the prescribed path. This path is entered in the control system of the vehicle as the so-called virtual guide, which describes the sequence of points, which in the given plane or in the given space form the desired path of the vehicle. This sequence of points is stored in a suitable data structure. By changing the contents of the corresponding data structure, the desired path of the vehicle can be adjusted or changed at any time. In doing so, the predictive control algorithm of the vehicle uses a time stamp attached to each position determination of the vehicle to continuously identify the delay resulting from the position determination by the sensor node. Thanks to the fact that the predictive control of the vehicle implicitly includes a compensator for this data transport delay, this data delay can also be compensated.

The second purpose is the continuous transmission of data about the current position of the respective sensor node, i.e., also of the vehicle on which the respective sensor node is mounted, via a wireless data connection to a common control system, which continuously monitors the position of the vehicles in a given space or in a given area and which, if necessary, performs coordination of movement of the vehicles and, if necessary, may remotely stop the vehicle or group of vehicles, adjust or change the path of the individual vehicle or group of vehicles. Due to the fact that time stamp is attached to each data item about the position of each vehicle, the common control system may evaluate both the speed and direction of the movement of each vehicle and perform adaptive compensation for the variable data delay that results from the position determination of the vehicle by the sensor node and the subsequent transmission of the data over the wireless network.

Description of the drawing

The invention is schematically represented in the drawing, wherein Fig. 1 shows an exemplary embodiment of a system according to the invention with a vehicle with a sensor node and a control computer in 3D side view, Fig. 1 a shows an exemplary embodiment of the system according to the invention with the vehicle and the sensor node and with the control computer in front view, Fig. 1 b shows an exemplary embodiment of the system according to the invention with the vehicle and the sensor node and with the control computer in top view and Fig. 2 is a plan view of an exemplary space in which the vehicle according to the invention moves, including the indication of the permissible edge paths of the vehicle.

Examples of embodiment of the invention

The invention will be described with reference to an exemplary embodiment of a method of navigating an automated guided vehicle for internal logistics and a system for its implementation.

The system for navigating an automated guided vehicle for internal logistics comprises at least one automated guided vehicle 1, also known by the abbreviation AGV, which is provided with an on-board control system 2 and a sensor node 3. The on-board control system 2 and the sensor node 3 are data linked.

The vehicle 1 further comprises a power source in the form of an unillustrated battery. In addition, it comprises a drive coupled to a movement system, e.g., a wheel or belt system. The wheel movement system comprises at least two steerable wheels 4 on opposite sides of the vehicle 1_, the belt movement system comprises two belts on opposite sides of the vehicle T

The movement system is coupled to a control system of controlling movement and parking of the vehicle 1 and a system for changing the direction of movement of the vehicle 1, e.g., in the case of the wheel movement system by turning the steerable wheels 4 or in the case of the belt movement system by means of the difference in the speed of movement of each belt.

The on-board control system 2 comprises means adapted for predictive control of the movement and parking of the vehicle 1 and for predictive control and change of the direction of movement of the vehicle 1, i.e., for predictive steering. The on-board control system 2 further comprises means adapted for wireless communication with a common control system 5. The means adapted for predictive control of the movement and parking of the vehicle 1 and control and change of the direction of the vehicle 1 movement, i.e., for predictive steering, preferably comprise local predictive control algorithm of the vehicle 1 with a mathematical model of the dynamic behaviour of the vehicle 1.

The sensor node 3 comprises an inertial measurement unit 30, a set of optical sensors 31 and a time unit 32.

In the illustrated exemplary embodiment, the set of optical sensors 31 includes four video-cameras 310 facing both sides of the vehicle 1, i.e., always two cameras 310 facing each side of the vehicle T Furthermore, the set of optical sensors 31 includes one forward-facing infrared depth camera 311 and one infrared laser projector 312 also facing forward, i.e., in the predominant direction P of travel of the vehicle . The optical sensors 31 are connected to the inertial measurement unit 30.

The inertial measurement unit 30 is adapted to process data obtained by the optical sensors 31, for mutual correlation of the data acquired by the individual optical sensors 31 and for obtaining information about the current position (x, y, z) of the sensor node 3 and about the rotation (a, b, c) of the sensor node 3, i.e., the angles of inclination, tilting and rotation of the sensor node 3.

The time unit 32 is adapted to issue a time stamp, as it is called in IT field, to be assigned to each data item about the position of the vehicle 1, wherein the time stamp describes the time point of data acquisition to determine the respective values of the current position (x, y, z) and current rotation (a, b, c) of the sensor node 3, i.e. of the vehicle 1.

The common control system 5 comprises means adapted to continuously monitor the position of the vehicle 1 or group of vehicles 1 in a specified space or in a specified area 6, for example, in Fig. 2 defined by the circumferential walls 60 and the edge line of the possible path 7 of the vehicle 1, means for adjustment or change of the path of movement of the vehicle 1 and/or group of vehicles 1, means for coordinating the movement of the group of vehicles 1, means for stopping the vehicle 1 and/or group of vehicles 1 etc., all this on the basis of the information about the current position (x, y, z) of the sensor node 3 of the respective vehicle 1, about the rotation (a, b, c) of the sensor node 3 of the respective vehicle 1 and about the time stamp assigned to the respective data about the current position and the rotation of the sensor node 3 of the respective vehicle 1. These means of the common control system 5 are formed, for example, by a common predictive control algorithm of the vehicle 1. and/or of group of vehicles T

The method of navigating an automated guided vehicle for internal logistics according to the invention is performed by running in a sensor node 3 software designed to detect the current positional coordinates of the sensor node 3, i.e., of the entire vehicle 1, by means of an inertial measurement unit 30 and a set of optical sensors 31. The sensor node 3 continuously acquires data (x, y, z, a, b, c) about the current position and about the current rotation of the sensor node 3, i.e., of the entire vehicle T Indication of time, i.e., a time stamp, is assigned by the time unit 32 to each data item (x, y, z, a, b, c) about the current position and about the current rotation of the sensor node 3, i.e., of the entire vehicle 1_.

Each data item (x, y, z, a, b, c) about the current position and about the current rotation of the sensor node 3 is transmitted to the on-board control system 2, which performs local predictive control of the vehicle 1 according to locally set local predictive control algorithm, according to which the drive of the vehicle 1 is controlled, as well as the direction of the vehicle 1. movement, steering of the vehicle 1, etc. Preferably, the local predictive control algorithm works with a fixed period and all steps of its activity are regularly repeated. The local predictive control algorithm processes the data (x, y, z, a, b, c) about the current position and about the current rotation of the sensor node 3 and determines the instantaneous distance of the vehicle 1 from the specified trajectory of movement of the vehicle 1. and also determines the deviation between the actual and desired angle of rotation of the vehicle T In addition, the local predictive control algorithm comprises a mathematical model of the dynamic behaviour of the vehicle 1, which is adapted to determine the need to change the direction of movement of the vehicle 1 in the future, e.g., by changing the angle of rotation of the steerable wheels 4. Optionally, the mathematical model of the dynamic behaviour of the vehicle 1 is also adapted to determine the speed of the vehicle 1 so that the actual trajectory of the vehicle 1 is as close as possible to the desired trajectory of the vehicle 1. The determination of the action commands for the elements performing the control of movement and direction of the vehicle 1 is carried out as a numerical optimization minimizing the criterion that summarizes the difference between the reference and the actual trajectories. In doing so, the limitations given by the dimensions and mechanical limitations of the vehicle, such as the maximum angle of rotation of the wheels are also taken into account. Of the specified action commands, the local predictive control algorithm always uses only the first information about the current position and rotation of the vehicle 1. valid for the current period, and in the next period the determination of action commands for powerful motion and direction control elements of the vehicle 1 is repeated, using updated information about the current position of the vehicle 1 from the sensor node 3. Since this information about the current position of the vehicle 1 from the sensor node 3 is always relative to the initial (starting) point where the vehicle 1 was started and where the determination of the current position and rotation of the vehicle 1 by the sensor node 3 was started, it is advantageous to coordinate the position of this starting point of the vehicle 1 , e.g., with the position according to the global coordinate system valid for the whole space and for all the vehicles 1 moving in this space. This can be done, for example, by switching on the vehicle 1 at a precisely defined location in a space whose global position coordinates are known.

The data (x, y, z, a, b, c) about the current position and about the current rotation of the sensor node 3 of the vehicle 1 are then transmitted via wireless communication, e.g., WiFi, to the common control system 5, e.g., created with the Node-Red programming tool, whereby the common control system 5 in the simplest case only logs the received messages about the position of the vehicle 1 to the database for possible further analysis and to display the position of the vehicle 1 on the map, see Fig. 2, wherein advanced control of the vehicle 1 and/or group of vehicles 1 is performed by common predictive control algorithm of the vehicle 1 and/or group of vehicles 1, which allows continuous monitoring of the position of the vehicle 1 or group of vehicles 1 in the specified space or in the specified area 6, predictive adjustment or change of the path of the vehicle 1 and/or group of vehicles 1, predictive coordination of movement of the group of vehicles 1, stopping the vehicle 1 and/or group of vehicles 1 , etc., all this on the basis of the information about the current position (x, y, z) of the sensor node 3 of each respective vehicle 1, about the rotation (a, b, c) of the sensor node 3 of each respective vehicle 1 and about the time stamp assigned to the respective data about the current position and rotation of the sensor node 3 of each respective vehicle T

The illustrated edge line of the possible path 7 of the vehicle 1 shown in Fig. 2 in essence indicates the minimum permissible distance of the vehicle 1 from the circumferential wall 60 of the area 6 for guaranteed and safe movement of the vehicle 1 in the area 6.

The invention is not limited to the embodiments explicitly described and illustrated herein but is applicable in other specific embodiments, considering the skills and abilities of a person skilled in the art.

Industrial applicability

The invention is applicable to be used especially for controlling automated guided vehicles in local logistics without the use of external guiding means.