Unmanned aerial vehicle cluster collaborative path planning method and system

Abstract

The invention discloses an unmanned aerial vehicle cluster collaborative path planning method and system, and belongs to the technical field of unmanned aerial vehicle cluster control. Accurate data modeling is realized for each unmanned aerial vehicle through acquisition and processing of unmanned aerial vehicle performance parameter data, and when task target area division and path planning are carried out on the unmanned aerial vehicles, the unmanned aerial vehicle cluster collaborative path planning is realized. The heterogeneous unmanned aerial vehicles are distributed and processed based on performance parameters of the unmanned aerial vehicles, so that accurate cooperative control of the heterogeneous unmanned aerial vehicles is realized; moreover, the corresponding regional space-time cost of each task target region is calculated, so that the accurate distribution of the task target regions is formed, the adaptation relationship between the performance of the unmanned aerial vehicle and the task target regions is ensured, and the task failure is avoided. Besides, through establishment and optimization of a path optimization objective function, a corresponding area access path is generated, and a corresponding spatial resultant force is calculated for each unmanned aerial vehicle, so that complete path planning of the unmanned aerial vehicles between the areas and in the areas is realized, and task execution efficiency, safety and reliability facing complex scenes are comprehensively improved.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) swarm control technology, specifically relating to a UAV swarm collaborative path planning method and system. Background Technology

[0002] With the rapid development of drone technology, drone swarms have shown significant advantages in performing large-area coverage tasks (such as search and rescue patrols and terrain mapping). However, existing drone swarm control technologies, particularly their path planning methods, are unable to achieve accurate and efficient planning and control when faced with complex multi-area, heterogeneous drones, and dynamic obstacle environments.

[0003] Specifically, most existing drone swarm control technologies are designed for planning and controlling the paths of homogeneous drone swarms within a single continuous area. However, in practical applications, when encountering emergencies requiring terrain mapping or personnel search and rescue, the available drones are often not all homogeneous, of the same model, or with the same performance. Existing drone swarm control technologies can only generate a unified path planning command based on the homogeneous drone swarm. This prevents the full utilization of the performance of heterogeneous drones. Furthermore, due to differences in endurance, speed, and payload capacity, the overall resource utilization of the swarm may be low, making it difficult to adapt to diverse mission requirements and even leading to mission failure. This seriously affects the deployment and execution of drone swarm missions.

[0004] Furthermore, in response to the problem of multi-area coverage, existing UAV swarm control technologies often treat area allocation and path generation separately, ignoring the time cost of task execution and the matching of UAV capabilities, resulting in problems such as long task completion time and uneven distribution of task areas. Moreover, although existing path planning technologies such as artificial potential field methods can achieve obstacle avoidance, they are prone to getting trapped in local minima and are difficult to generate smooth paths that match the actual performance parameters of UAVs in complex post-disaster search and rescue environments.

[0005] As mentioned above, how to provide a method and system for collaborative path planning of UAV swarms that can accurately coordinate and control heterogeneous UAVs and realize integrated path planning in multiple regions has become an urgent problem to be solved in this field. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for collaborative path planning of unmanned aerial vehicle (UAV) swarms, in order to solve the above-mentioned problems existing in the prior art.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for collaborative path planning of unmanned aerial vehicle (UAV) swarms, comprising: The system acquires UAV mission environment data and UAV performance parameter data, and preprocesses and calculates secondary parameter data on the UAV mission environment data and UAV performance parameter data to obtain UAV cluster mission configuration data. The UAV cluster mission configuration data includes multiple mission target areas and UAV performance parameters of each UAV in the UAV cluster. Based on the UAV cluster task configuration data, the corresponding regional spatiotemporal cost is calculated for each task target area. Based on the regional spatiotemporal cost of each task target area and the UAV performance parameters of each UAV, a task target area allocation table is generated. The task target area allocation table is used to represent the allocation result of the task target area corresponding to each UAV, and each UAV is allocated at least one task target area. With the goal of minimizing the access distance of a single UAV and the access time of a UAV cluster, a path optimization objective function is established for each UAV based on the task target area allocation table to calculate the area access path for each UAV. Based on the UAV cluster task configuration data, the corresponding spatial force is calculated for each UAV to generate a corresponding path point sequence for each UAV using the spatial force. The access path for each UAV is then formed based on the path point sequence. The access paths of each drone are integrated to form a drone cluster collaborative path planning scheme. Based on the drone cluster collaborative path planning scheme, corresponding access control commands are issued to each drone to control each drone to perform access tasks according to the access control commands.

[0008] In one possible design, UAV mission environment data and UAV performance parameter data are acquired, and the UAV mission environment data and UAV performance parameter data are preprocessed to obtain UAV swarm mission configuration data, including: The drone management platform obtains the drone cluster task description, extracts the geographic area information to be accessed from the drone cluster task description, generates multiple task target areas based on the geographic area information to be accessed, and obtains the regional center coordinates, regional boundary vertex coordinates, regional area, regional geometric shape description and regional terrain elevation data of each task target area through the drone management platform to form the task target area information of each task target area. The takeoff base coordinates and return base coordinates are extracted from the drone cluster mission description, and the takeoff base coordinates are used as the starting point of the path planning and the return base coordinates are used as the ending point of the path planning to form initial path planning information; According to the description of the UAV cluster mission, multiple environmental obstacles are extracted from each of the mission target areas, and the obstacle type, obstacle center coordinates and maximum influence radius of each environmental obstacle are obtained through the UAV management platform to form obstacle information for each environmental obstacle; The mission target area information, the initial path planning information, and the obstacle information of each environmental obstacle are integrated into UAV mission environment data. By using the drone management platform, the flight capability parameters, endurance parameters, mission payload parameters, and kinematic parameters of each drone in the drone cluster are obtained to form the drone performance parameters of each drone. By using the drone cluster task description, drone task constraints and drone safety constraints are obtained to form drone constraint data. Obtain a preset task configuration data format, and perform data formatting processing on the UAV task environment data, the UAV performance parameter data and the UAV constraint data through the task configuration data format. Perform coordinate unification processing on the UAV task environment data and the UAV performance parameter data to obtain preprocessed UAV task environment data and preprocessed UAV performance parameter data. Based on the UAV mission environment data and the UAV performance parameter data, the regional distance between each mission target area, the flight time of each UAV between each mission target area, and the scanning time of each UAV to each mission target area are calculated to form secondary parameter data; The preprocessed UAV mission environment data, the preprocessed UAV performance parameter data, and the secondary parameter data are integrated to form UAV swarm mission configuration data.

[0009] In one possible design, based on the drone swarm mission configuration data, the corresponding regional spatiotemporal cost for each mission target area is calculated, including: The spatial neighborhood threshold and the temporal neighborhood threshold are calculated using the drone cluster task configuration data, and based on the spatial neighborhood threshold and the temporal neighborhood threshold, the corresponding spatiotemporal neighborhood task area is selected for each of the task target areas. For each of the task target regions, the scanning time cost between each of the task target regions and the corresponding spatiotemporal neighbor task regions is calculated using the following formula (1): (1) in, and All of these are indexes of the target region of the task. The total number of the target areas for the task. It is a binary function. Indicates the target area of ​​the task With the target area of ​​the task The distance between the regions, Indicates that the drone is in the mission target area With the target area of ​​the task Flight time between Indicates that the drone is in the mission target area Scanning time within, Indicates that the drone is in the mission target area Scanning time within, This represents the spatial neighborhood threshold. This represents the time neighborhood threshold. Indicates the target area of ​​the task The scanning time cost between the corresponding spatiotemporal neighboring task regions; The corresponding scanning time cost is calculated for each of the mission target areas, and the flight space cost corresponding to each of the mission target areas is calculated based on the scanning time cost corresponding to each of the mission target areas. Based on the scanning time cost and flight space cost of each UAV in each of the aforementioned mission target areas, the regional spatiotemporal cost corresponding to each mission target area is calculated using the following formula (2): (2) in, Indicates the target area of ​​the mission The cost of flight space, Indicates the target area of ​​the mission Regional time and space costs.

[0010] In one possible design, a task target area allocation table is generated based on the spatiotemporal cost of each task target area and the performance parameters of each UAV, including: Based on the regional spatiotemporal cost of each of the task target regions, the task target regions are arranged in descending order of regional spatiotemporal cost to form a task target region sequence. Based on the drone swarm mission configuration data, performance coefficients are generated for each drone, and the drones are arranged in descending order of performance coefficients to form a drone swarm sequence. Based on the mission target area sequence and the UAV cluster sequence, a corresponding area allocation relationship is established between each UAV and each mission target area. The area allocation relationships corresponding to each UAV are integrated to generate a mission target area allocation table.

[0011] In one possible design, based on the mission target region sequence and the UAV swarm sequence, a corresponding region allocation relationship is established between each UAV and each mission target region. The region allocation relationships corresponding to each UAV are then integrated to generate a mission target region allocation table, including: The number of drones in the drone swarm sequence and the number of mission target areas in the mission target area sequence are compared to obtain the comparison results. When the comparison result indicates that the number of drones in the drone swarm sequence is greater than the number of mission target areas in the mission target area sequence, the method includes: The last drone in the drone cluster sequence is removed sequentially until the number of drones in the drone cluster sequence is equal to the number of task target areas in the task target area sequence. Each task target region in the task target region sequence is assigned one-to-one to each drone in the drone cluster sequence after the removal is completed, forming a region allocation relationship for each drone. The region allocation relationships for each drone are then integrated to generate a task target region allocation table. When the comparison result indicates that the number of drones in the drone swarm sequence is no more than the number of mission target areas in the mission target area sequence, the method includes: According to the mission target area sequence, the mission target area is allocated to each drone in the drone cluster sequence to form a core scanning area corresponding to each drone. The core scanning area corresponding to each drone is used as the first element to establish a scanning area set for each drone. According to the mission target area sequence, the scanning area sets corresponding to each UAV are arranged in order to form the UAV scanning area sequence, and according to the mission target area sequence, the unassigned mission target areas are arranged in reverse order to form the unassigned area sequence. The unallocated region sequence is filled into the scan region set of the UAV scan region sequence in the corresponding order, sorted after the first element to form a pre-filled scan region set, and the total access time of the pre-filled scan region set is calculated, wherein the total access time includes the scanning time of the UAV scanning each task target region in the pre-filled scan region set and the shortest time of the UAV flying between each task target region in the pre-filled scan region set. The flight endurance parameters of each drone are extracted from the drone cluster task configuration data. The total flight endurance of each drone is calculated based on the flight endurance parameters of each drone. The total flight endurance of each drone is used to verify the total access time of the pre-filled scan area set corresponding to each drone, and the verification results are obtained. If the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone does not exceed the total flight time of the current drone, then this filling is performed to update the pre-filled scan area set to a scan area set. The unallocated areas in the unallocated area sequence are then used to fill the scan area set in order to form a pre-filled scan area set again. The verifiable verification is then performed again until the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone exceeds the total flight time of the current drone. If the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone exceeds the total flight time of the current drone, then the filling will not be performed, the scan area set corresponding to the current drone will be used as the final scan area set, and the scan area set corresponding to the current drone will be removed from the drone scan area sequence to complete the update of the drone scan area sequence. The task target regions in the unallocated region sequence are sequentially filled into the UAV scanning region sequence until the UAV scanning region sequence is empty. The final scanning region set corresponding to each UAV is obtained. The final scanning region set corresponding to each UAV is used as the region allocation relationship corresponding to each UAV. The region allocation relationships corresponding to each UAV are integrated to generate a task target region allocation table.

[0012] In one possible design, with the objectives of minimizing the individual drone access distance and the drone swarm access time, a path optimization objective function is established for each drone based on the task objective area allocation table to calculate the area access path for each drone, including: Extract the path planning start point and path planning end point from the drone cluster mission configuration data; Based on the task target area allocation table, the path planning start point, and the path planning end point, a corresponding access distance minimization solution function is established for each UAV. The access distance is the sum of the flight distance between the UAV from the path planning start point to the corresponding initial task target area in the task target area allocation table, the flight distance between each task target area in the task target area allocation table, and the flight distance between the UAV from the corresponding final task target area in the task target area allocation table to the path planning end point. Based on the task target area allocation table, the path planning start point, and the path planning end point, a corresponding access time maximization function is established for each UAV. The access time is the sum of the UAV's flight time from the path planning start point to the corresponding initial task target area in the task target area allocation table, the UAV's flight time between each task target area in the task target area allocation table, the UAV's scan time within each task target area in the task target area allocation table, and the UAV's flight time from the corresponding final task target area in the task target area allocation table to the path planning end point. The function for minimizing the access distance of each drone is used as the optimization function for the shortest access distance of each individual drone. The minimum value of the function for maximizing the access time of each drone is obtained to get the optimization function for the shortest access time of the drone cluster. Obtain a preset target weight, and sum the access distance minimization solution function and the drone cluster access time shortest optimization function based on the target weight to form a path optimization objective function for each drone. For each UAV, an initial path solution is generated based on the path optimization objective function, so as to form a path solution space for each UAV based on each path solution. Based on the genetic algorithm, the path solution space corresponding to each UAV is optimized to obtain the optimal path solution for each UAV, and the optimal path solution is used as the area access path for each UAV.

[0013] In one possible design, a genetic algorithm is used to optimize the path solution space for each UAV, obtaining the optimal path solution for each UAV. This optimal path solution is then used as the area access path for each UAV, including: In the path solution space corresponding to each UAV, a path solution is taken as a chromosome individual to form an initial chromosome population. The path solutions corresponding to the order of the task target region in the task target region allocation table are taken as initial elite chromosome individuals and added to the initial chromosome population. The initial chromosome population is taken as the parent chromosome population. Based on the function value of the path optimization objective function corresponding to each individual chromosome, the fitness of each individual chromosome in the parent chromosome population is calculated using the following formula (3): (3) in, This is the index of the individual chromosome. Chromosomal individuals The corresponding function value of the path optimization objective function, For division by zero parameters, Chromosomal individuals Corresponding fitness; A preset elite individual retention ratio is obtained. Based on the elite individual retention ratio and the number of chromosome individuals in the parent chromosome population, the number of chromosomes to be retained is calculated. The multiple chromosome individuals with the highest fitness are selected as elite chromosome individuals and inherited to the offspring chromosome population. The number of elite chromosome individuals is determined by the number of chromosomes to be retained. A preset parent candidate ratio is obtained. Based on the parent candidate ratio and the number of chromosome individuals in the parent chromosome population, the number of parent candidates is calculated. Using a roulette wheel method, multiple chromosome individuals are selected from the initial chromosome population as parent candidate crossover chromosome individuals. The number of parent candidate crossover chromosome individuals is determined by the number of parent candidates, and the sum of the parent candidate ratio and the elite individual retention ratio is 1. Obtain a preset crossover inheritance probability, perform partial mapping crossover on each of the parent candidate crossover chromosome individuals based on the crossover inheritance probability to obtain crossover offspring chromosomes, and supplement the crossover offspring chromosomes to the pre-offspring chromosome population; A preset mutation probability is obtained, and a mutation operation is performed on the offspring chromosome population based on the mutation probability to obtain an offspring chromosome population. The mutation operation includes exchange mutation, reverse mutation, and insertion mutation. Obtain the preset maximum number of generations, update the offspring chromosome population to the parent chromosome population, and perform multiple rounds of genetic iteration until the number of genetic iterations reaches the maximum number of generations, complete the iteration, and select the chromosome individual with the highest fitness as the optimal chromosome individual from the genetic iteration process. The optimal path solution corresponding to the optimal chromosome individual is used as the optimal path solution for the UAV. The corresponding optimal path solution is calculated for each UAV, and the optimal path solution corresponding to each UAV is used as the area access path for each UAV.

[0014] In one possible design, based on the drone swarm mission configuration data, the corresponding spatial resultant force for each drone is calculated. This spatial resultant force is then used to generate a corresponding pathpoint sequence for each drone. Based on this pathpoint sequence, an access path for each drone is formed, including: Based on the aforementioned UAV cluster mission configuration data, a three-dimensional simulation mission environment map is constructed. In the three-dimensional simulation mission environment map, the regional center coordinates of each mission target area are extracted, and based on the regional access paths of each UAV, the regional center coordinates of each mission target area are marked as gravity points. The area access path of each UAV is divided into multiple access path intervals according to each gravity point, and each gravity point is set as the interval endpoint of each access path interval. In the three-dimensional simulation mission environment map, the gravitational force exerted on the UAV by the corresponding endpoint of each access path segment is calculated. In the three-dimensional simulation task environment map, obstacle information of each task target area is extracted, and according to the obstacle information, each obstacle is placed in each access path interval, so that the obstacle center coordinates of the obstacle in the corresponding access path interval are used as the repulsion point. In each of the access path intervals, the repulsive force generated by the corresponding repulsive point on the drone is calculated; Obtain a preset tangential perturbation gain coefficient. In each segment of the access path interval, when the resultant force of the attraction and the repulsion is 0, generate a corresponding tangential perturbation force for the attraction and the repulsion through the tangential perturbation gain coefficient. By integrating the gravitational force, the repulsive force, and the tangential perturbation force, the spatial resultant force of each UAV at each moment within each segment of the access path is generated; The performance parameters of each drone are extracted from the drone swarm mission configuration data, and the maximum flight speed of each drone is calculated based on these parameters. Based on the spatial resultant force of each UAV at each moment within each access path interval, the movement direction of each UAV at each moment is generated in the three-dimensional simulation task environment map. Using the movement direction of each UAV at each moment and the maximum flight speed of each UAV, the path points of each UAV at each moment are generated. For each drone, the path points in each access path interval and the end point of each access path interval are sequentially integrated to generate a path point sequence corresponding to each drone. The path point sequence corresponding to each UAV is smoothed to obtain the access path of each UAV.

[0015] In one possible design, the access paths of each drone are integrated to form a drone swarm collaborative path planning scheme. Based on the drone swarm collaborative path planning scheme, corresponding access control commands are issued to each drone to control each drone to perform access tasks according to the access control commands, including: Extract drone constraint data from the drone cluster mission configuration data; Based on the task target area allocation table, the access paths of each UAV are obtained sequentially, and the UAV constraint data is used to determine whether there is a collision risk in the access paths of each UAV. If not, the access paths of each UAV are integrated to form a UAV cluster collaborative path planning scheme. Based on the UAV cluster collaborative path planning scheme, corresponding access control commands are issued to each UAV, and a synchronous take-off trigger command is added to the access control command corresponding to each UAV to control each UAV to take off synchronously and execute the access task according to the corresponding access control command. If so, then the access paths with collision risks are selected, and corresponding access control instructions are generated for each access path with collision risks. An asynchronous take-off instruction is added to the access control instructions corresponding to each access path with collision risks. For each access path without collision risks, corresponding access control instructions are generated, and a synchronous take-off trigger instruction is added to the access control instructions corresponding to each access path without collision risks, so as to control each UAV to take off synchronously / asynchronously and execute the access task according to the corresponding access control instructions.

[0016] Secondly, the present invention provides a collaborative path planning system for unmanned aerial vehicle (UAV) swarms, comprising: The data acquisition unit is used to acquire UAV mission environment data and UAV performance parameter data, and to preprocess the UAV mission environment data and the UAV performance parameter data and calculate secondary parameter data to obtain UAV cluster mission configuration data. The UAV cluster mission configuration data includes multiple mission target areas and UAV performance parameters of each UAV in the UAV cluster. The region allocation unit is used to calculate the corresponding regional spatiotemporal cost for each task target region based on the UAV cluster task configuration data, and generate a task target region allocation table based on the regional spatiotemporal cost of each task target region and the UAV performance parameters of each UAV. The task target region allocation table is used to represent the allocation result of the task target region corresponding to each UAV, and each UAV is allocated at least one task target region. The path generation unit is used to optimize the path for each UAV based on the task target area allocation table, with the goal of minimizing the access distance of a single UAV and the access time of the UAV cluster. It calculates the area access path for each UAV and calculates the corresponding spatial force for each UAV based on the UAV cluster task configuration data. It then uses the spatial force to generate a corresponding path point sequence for each UAV and forms the access path for each UAV based on the path point sequence. The command issuing unit integrates the access paths of various UAVs to form a UAV swarm collaborative path planning scheme. Based on the UAV swarm collaborative path planning scheme, it issues corresponding access control commands to each UAV to control each UAV to perform access tasks according to the access control commands. Thirdly, the present invention provides an electronic device comprising a memory, a processor, and a transceiver connected in sequence and communication, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the UAV swarm cooperative path planning method as described in the first aspect or any possible design of the first aspect.

[0017] Fourthly, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, perform the UAV swarm cooperative path planning method described in the first aspect or any possible design of the first aspect.

[0018] Fifthly, the present invention provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the UAV swarm cooperative path planning method as described in the first aspect or any possible design of the first aspect.

[0019] Beneficial Effects: This invention provides a method and system for collaborative path planning in a drone swarm, comprising: First, acquiring drone mission environment data and drone performance parameter data, and preprocessing and calculating secondary parameter data on the drone mission environment data and drone performance parameter data to obtain drone swarm mission configuration data, wherein the drone swarm mission configuration data includes multiple mission target areas and drone performance parameters of each drone in the drone swarm; Second, based on the drone swarm mission configuration data, calculating the corresponding regional spatiotemporal cost for each mission target area, and generating a mission target area allocation table based on the regional spatiotemporal cost of each mission target area and the drone performance parameters of each drone, wherein the mission target area allocation table is used to represent the mission corresponding to each drone. The target area allocation results are determined, with each drone assigned at least one task target area. Then, with the objectives of minimizing the individual drone's access distance and the drone swarm's access time, a path optimization objective function is established for each drone based on the task target area allocation table. This function calculates the area access path for each drone. Furthermore, based on the drone swarm task configuration data, the corresponding spatial force is calculated for each drone. This spatial force is used to generate a corresponding path point sequence for each drone, forming the access path for each drone. Finally, the access paths of each drone are integrated to form a drone swarm collaborative path planning scheme. Based on this scheme, corresponding access control commands are issued to each drone to control them to execute access tasks according to the commands. By acquiring and processing UAV performance parameter data, accurate data modeling is achieved for each UAV. When dividing the mission target area and planning the path, allocation and processing are based on its performance parameters, realizing accurate collaborative control of various heterogeneous UAVs. Furthermore, the corresponding spatiotemporal cost of each mission target area is calculated to form an accurate allocation of the mission target area, ensuring the adaptation relationship between UAV performance and mission target area and avoiding mission failure. In addition, by establishing and optimizing the path optimization objective function, corresponding area access paths are generated, and the corresponding spatial resultant force is calculated for each UAV to achieve complete path planning between and within regions, comprehensively improving the mission execution efficiency, safety, and reliability in complex scenarios. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the UAV swarm cooperative path planning method provided in an embodiment of the present invention. Figure 2 This is a functional structure diagram of the UAV swarm collaborative path planning system provided in an embodiment of the present invention; Figure 3This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.

[0022] It should be understood that although the terms first, second, etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, without departing from the scope of the exemplary embodiments of the invention.

[0023] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.

[0024] Example: like Figure 1 As shown, the first aspect of this embodiment provides a method for collaborative path planning of unmanned aerial vehicle (UAV) swarms, which may include, but is not limited to, the following steps: S1. Acquire UAV mission environment data and UAV performance parameter data, and preprocess the UAV mission environment data and the UAV performance parameter data and calculate secondary parameter data to obtain UAV cluster mission configuration data, wherein the UAV cluster mission configuration data includes multiple mission target areas and UAV performance parameters of each UAV in the UAV cluster. In one possible implementation, step S1 involves acquiring UAV mission environment data and UAV performance parameter data, and preprocessing the UAV mission environment data and UAV performance parameter data to obtain UAV cluster mission configuration data. This can be, but is not limited to, decomposed into the following steps S11-S19, specifically including: S11. Obtain the drone cluster task description through the drone management platform, extract the geographic area information to be accessed from the drone cluster task description, generate multiple task target areas based on the geographic area information to be accessed, and obtain the regional center coordinates, regional boundary vertex coordinates, regional area, regional geometric shape description and regional terrain elevation data of each task target area through the drone management platform to form the task target area information of each task target area. S12. Extract the takeoff base coordinates and return base coordinates from the UAV cluster mission description, and use the takeoff base coordinates as the starting point of the path planning and the return base coordinates as the ending point of the path planning to form initial path planning information; S13. Based on the UAV cluster mission description, extract multiple environmental obstacles from each of the mission target areas, and obtain the obstacle type, obstacle center coordinates and maximum influence radius of each environmental obstacle through the UAV management platform to form obstacle information for each environmental obstacle; S14. Integrate the mission target area information of each mission target area, the initial path planning information, and the obstacle information of each environmental obstacle into UAV mission environment data; S15. Through the drone management platform, obtain the flight capability parameters, endurance parameters, mission payload parameters and kinematic parameters of each drone in the drone cluster to form the drone performance parameters of each drone. S16. Obtain drone task constraints and drone safety constraints through the drone cluster task description to form drone constraint data; S17. Obtain a preset task configuration data format, and perform data formatting processing on the UAV task environment data, the UAV performance parameter data and the UAV constraint data through the task configuration data format, and perform coordinate unification processing on the UAV task environment data and the UAV performance parameter data to obtain preprocessed UAV task environment data and preprocessed UAV performance parameter data. S18. Based on the UAV mission environment data and the UAV performance parameter data, calculate the regional distance between each mission target area, the flight time of each UAV between each mission target area, and the scanning time of each UAV to each mission target area to form secondary parameter data; S19. Integrate the preprocessed UAV mission environment data, the preprocessed UAV performance parameter data, and the secondary parameter data to form UAV cluster mission configuration data.

[0025] It should be noted that in the UAV swarm collaborative path planning method provided in this embodiment, the flight capability parameters include at least the UAV's maximum cruising speed, economic cruising speed, maximum acceleration, and maximum angular velocity; the endurance parameters include at least the UAV's maximum continuous flight time and / or maximum flight distance; the mission payload parameters include at least the UAV's effective scanning radius and payload capacity limit; and the kinematic parameters include at least the UAV's mass, inertia matrix, and coefficient matrix. Furthermore, in one possible but not limited embodiment, the UAV mission constraints include at least the UAV scan mission coverage requirement (e.g., 100% coverage of the mission target area), total UAV mission time limit requirement, UAV area scan order requirement, and UAV path overlap constraint; the UAV safety constraints include at least the minimum safe distance between UAVs, the coordinate range of the no-fly zone, and the maximum permissible flight altitude. Introducing UAV constraint data can effectively avoid collision risks between UAV swarms and optimize and investigate duplicate scans, duplicate flights, and missed area detections, thereby improving the safety and reliability of the final generated access path and ensuring the scan mission coverage and mission efficiency of the UAV swarm.

[0026] S2. Based on the UAV cluster task configuration data, calculate the corresponding regional spatiotemporal cost for each task target area. Based on the regional spatiotemporal cost of each task target area and the UAV performance parameters of each UAV, generate a task target area allocation table. The task target area allocation table is used to represent the allocation result of the task target area corresponding to each UAV, and each UAV is allocated at least one task target area. In one possible implementation, step S2, based on the UAV cluster mission configuration data, calculates the corresponding regional spatiotemporal cost for each mission target area. This can be decomposed into, but is not limited to, the following steps S21-S24, specifically including: S21. Calculate the spatial neighborhood threshold and the temporal neighborhood threshold using the drone cluster task configuration data, and select the corresponding spatiotemporal neighborhood task area for each of the task target areas based on the spatial neighborhood threshold and the temporal neighborhood threshold. S22. For each of the task target regions, the scanning time cost between each of the task target regions and the corresponding spatiotemporal neighbor task regions is calculated using the following formula (1): (1) in, and All of these are indexes of the target region of the task. The total number of the target areas for the task. It is a binary function. Indicates the target area of ​​the task With the target area of ​​the task The distance between the regions, Indicates that the drone is in the mission target area With the target area of ​​the task Flight time between Indicates that the drone is in the mission target area Scanning time within, Indicates that the drone is in the mission target area Scanning time within, This represents the spatial neighborhood threshold. This represents the time neighborhood threshold. Indicates the target area of ​​the task The scanning time cost between the corresponding spatiotemporal neighboring task regions; S23. Calculate the corresponding scanning time cost for each of the mission target areas, and calculate the corresponding flight space cost for each of the mission target areas based on the scanning time cost for each of the mission target areas; S24. Based on the scanning time cost and flight space cost of each UAV in each of the mission target areas, the regional spatiotemporal cost corresponding to each mission target area is calculated using the following formula (2): (2) in, Indicates the target area of ​​the mission The cost of flight space, Indicates the target area of ​​the mission Regional time and space costs.

[0027] It should be noted that in the UAV swarm collaborative path planning method provided in this embodiment, the spatial neighborhood threshold is calculated by calculating the standard deviation and average value of the distance between the regional center coordinates of all task target areas (i.e., the regional spacing between the task target areas), and then using the kernel density estimation method. The temporal neighborhood threshold is calculated by calculating the standard deviation and average value of the average flight time of all UAVs between all task target areas (i.e., the flight time of each UAV between each task target area), and then using the kernel density estimation method.

[0028] Specifically, for a given task target area, each task target area that has a spatial distance between itself and other task target areas not exceeding the spatial neighborhood threshold, and whose flight time between each UAV and other task target areas does not exceed the temporal neighborhood threshold, is selected as the spatiotemporal neighborhood task area of ​​this task target area.

[0029] In addition, scan time cost is used to characterize the expected time cost of a mission target region to perform a mission within its corresponding spatiotemporal neighborhood mission region. For mission target regions where the scan time cost is not the highest, the flight space cost is the minimum regional distance between the current mission target region and each mission target region with a higher scan time cost. For mission target regions with the highest scan time cost, the flight space cost is the maximum regional distance between the current mission target region and all mission target regions.

[0030] In one possible implementation, step S2, generating a task target area allocation table based on the spatiotemporal cost of each task target area and the performance parameters of each UAV, can be, but is not limited to, decomposed into the following steps S25-S27, specifically including: S25. Based on the regional spatiotemporal cost of each of the task target regions, arrange the task target regions in descending order of regional spatiotemporal cost to form a task target region sequence; S26. Based on the UAV cluster task configuration data, generate performance coefficients for each UAV and arrange them into a UAV cluster sequence in descending order of performance coefficients; S27. Based on the mission target area sequence and the UAV cluster sequence, establish corresponding area allocation relationships between each UAV and each mission target area, integrate the area allocation relationships corresponding to each UAV, and generate a mission target area allocation table.

[0031] In one possible implementation, step S27 involves establishing corresponding region allocation relationships between each drone and each task target region based on the task target region sequence and the drone cluster sequence, integrating the region allocation relationships corresponding to each drone, and generating a task target region allocation table. This can be decomposed into, but is not limited to, the following steps S271-S272, specifically including: S271. Compare the number of drones in the drone cluster sequence with the number of task target areas in the task target area sequence to obtain a comparison result; Wherein, when the comparison result is that the number of drones in the drone cluster sequence is greater than the number of task target areas in the task target area sequence, the method of step S272 may include, but is not limited to, the following steps S272a-S272b, specifically: S272a. Remove the last drone from the drone cluster sequence one by one until the number of drones in the drone cluster sequence is equal to the number of mission target areas in the mission target area sequence; S272b. Assign each task target region in the task target region sequence to each drone in the drone cluster sequence after the removal is completed, forming a region allocation relationship for each drone, and integrate the region allocation relationships for each drone to generate a task target region allocation table. Wherein, when the comparison result is that the number of drones in the drone cluster sequence is no more than the number of task target areas in the task target area sequence, the method of step S272 may include, but is not limited to, the following steps S272c-S272i, specifically: S272c. According to the mission target area sequence, allocate mission target areas to each UAV in the UAV cluster sequence one-to-one to form a core scanning area corresponding to each UAV, and use the core scanning area corresponding to each UAV as the first element to establish a scanning area set for each UAV. S272d. According to the mission target area sequence, the scanning area sets corresponding to each UAV are arranged in order to form the UAV scanning area sequence, and according to the mission target area sequence, the unassigned mission target areas are arranged in reverse order to form the unassigned area sequence; S272e. The unallocated region sequence is filled into the scan region set of the UAV scan region sequence in the order corresponding to the first element to form a pre-filled scan region set, and the total access time of the pre-filled scan region set is calculated, wherein the total access time includes the scanning time of the UAV scanning each task target region in the pre-filled scan region set and the shortest time of the UAV flying between each task target region in the pre-filled scan region set; S272f. Extract the endurance parameters of each UAV from the UAV cluster task configuration data, calculate the total endurance time of each UAV based on the endurance parameters of each UAV, and use the total endurance time of each UAV to perform feasible verification on the total access time of the pre-filled scan area set corresponding to each UAV, and obtain the verification results. S272g. If the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone does not exceed the total flight time of the current drone, then this filling is performed to update the pre-filled scan area set to a scan area set, and the unallocated areas in the unallocated area sequence are sequentially filled to the scan area set to form a pre-filled scan area set again, and the verifiable verification is performed again until the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone exceeds the total flight time of the current drone; S272h. If the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone exceeds the total flight time of the current drone, then this filling is not performed, the scan area set corresponding to the current drone is taken as the final scan area set, and the scan area set corresponding to the current drone is removed from the drone scan area sequence to complete the update of the drone scan area sequence. S272i. Fill each task target region in the unallocated region sequence into the UAV scanning region sequence in order until the UAV scanning region sequence is empty, obtain the final scanning region set corresponding to each UAV, use the final scanning region set corresponding to each UAV as the region allocation relationship corresponding to each UAV, and integrate the region allocation relationships corresponding to each UAV to generate a task target region allocation table.

[0032] In a possible implementation, in order to facilitate the determination of the initial elite chromosome individuals in step S3 (i.e., step S371, to improve the initial excellence of the initial chromosome population) and the sequential checking of each path in step S4 (i.e., step S42, to improve the efficiency of instruction generation), when generating the task target area allocation table, the order of the task target area allocation table can be arranged according to the UAV cluster sequence, and each task target area can be sorted according to the element arrangement order in the final scan area set corresponding to each UAV.

[0033] In specific application scenarios, by simultaneously considering the spatial distance between regions and the time cost of flight transfer, and combining the heterogeneous performance of UAVs to adaptively allocate core scanning regions and evenly allocate the remaining unallocated regions, optimal matching between each task target region and each UAV resource is achieved. Simulation experiments were conducted on digital twins of multiple search and rescue environments (using homogeneous UAV swarms to avoid introducing performance parameter variables). The results showed that the UAV swarm collaborative path planning method provided in this embodiment can effectively shorten the total flight path of the UAV swarm by more than 10% compared to the traditional nearest-neighbor clustering region allocation scheme. Furthermore, in multiple experiments, the overall average swarm scanning task completion time was shortened by nearly 5% compared to the traditional nearest-neighbor clustering region allocation scheme. This indicates that the region allocation method in this embodiment can significantly improve the overall task execution efficiency of the UAV swarm.

[0034] S3. With the goal of minimizing the access distance of a single UAV and the access time of a UAV cluster, a path optimization objective function is established for each UAV based on the task target area allocation table to calculate the area access path for each UAV. Based on the UAV cluster task configuration data, the corresponding spatial force is calculated for each UAV to generate a corresponding path point sequence for each UAV using the spatial force. The access path for each UAV is formed based on the path point sequence. In one possible implementation, step S3, with the objectives of minimizing the individual drone access distance and minimizing the drone cluster access time, establishes a path optimization objective function for each drone based on the task target area allocation table to calculate the area access path for each drone. This can be decomposed into, but is not limited to, the following steps S31-S37, specifically including: S31. Extract the path planning start point and path planning end point from the drone cluster mission configuration data; S32. Based on the task target area allocation table, the path planning start point, and the path planning end point, establish a corresponding access distance minimization solution function for each UAV, wherein the access distance is the sum of the flight distance between the UAV from the path planning start point to the corresponding initial task target area in the task target area allocation table, the flight distance between each task target area in the task target area allocation table, and the flight distance between the UAV from the corresponding final task target area in the task target area allocation table to the path planning end point; S33. Based on the task target area allocation table, the path planning start point, and the path planning end point, establish a corresponding access time maximization function for each UAV, wherein the access time is the sum of the UAV's flight time from the path planning start point to the corresponding initial task target area in the task target area allocation table, the UAV's flight time between each task target area in the task target area allocation table, the UAV's scan time within each task target area in the task target area allocation table, and the UAV's flight time from the corresponding final task target area in the task target area allocation table to the path planning end point. S34. The function for minimizing the access distance of each UAV is used as the optimization function for the shortest access distance of each UAV individual. The minimum value of the function for maximizing the access time of each UAV is obtained to get the optimization function for the shortest access time of the UAV cluster. S35. Obtain the preset target weights, and sum the access distance minimization solution function and the drone cluster access time shortest optimization function based on the target weights to form the path optimization objective function for each drone respectively; S36. For each UAV, generate a corresponding initial path solution based on the path optimization objective function, so as to form a path solution space for each UAV based on each path solution; S37. Based on the genetic algorithm, the path solution space corresponding to each UAV is optimized to obtain the optimal path solution corresponding to each UAV, and the optimal path solution is used as the area access path of each UAV.

[0035] In one possible implementation, step S37 involves optimizing the path solution space for each UAV based on a genetic algorithm to obtain the optimal path solution for each UAV, which is then used as the area access path for each UAV. This step can be decomposed into, but is not limited to, the following steps S371-S378, specifically including: S371. In the path solution space corresponding to each UAV, take a path solution as a chromosome individual to form an initial chromosome population, and take the path solutions corresponding to the order of the task target region in the task target region allocation table as initial elite chromosome individuals, add them to the initial chromosome population, and take the initial chromosome population as the parent chromosome population. S372. Based on the function value of the path optimization objective function corresponding to each individual chromosome, the fitness of each individual chromosome in the parent chromosome population is calculated using the following formula (3): (3) in, This is the index of the individual chromosome. Chromosomal individuals The corresponding function value of the path optimization objective function, For division by zero parameters, Chromosomal individuals Corresponding fitness; S373. Obtain a preset elite individual retention ratio, calculate the number of chromosomes to be retained based on the elite individual retention ratio and the number of chromosomes in the parent chromosome population, select the multiple chromosomes with the highest fitness as elite chromosomes, and pass them on to the offspring chromosome population, wherein the number of elite chromosomes is determined by the number of chromosomes to be retained; S374. Obtain a preset parent candidate ratio, calculate the number of parent candidates based on the parent candidate ratio and the number of chromosome individuals in the parent chromosome population, and select multiple chromosome individuals from the initial chromosome population as parent candidate crossover chromosome individuals using a roulette wheel method. The number of parent candidate crossover chromosome individuals is determined by the number of parent candidates, and the sum of the parent candidate ratio and the elite individual retention ratio is 1. S375. Obtain a preset crossover inheritance probability, perform partial mapping crossover on each of the parent candidate crossover chromosome individuals based on the crossover inheritance probability to obtain crossover offspring chromosomes, and supplement the crossover offspring chromosomes to the pre-offspring chromosome population. S376. Obtain a preset mutation probability, and perform mutation operations on the offspring chromosome population based on the mutation probability to obtain an offspring chromosome population, wherein the mutation operations include exchange mutation, reverse mutation and insertion mutation; S377. Obtain the preset maximum number of generations, update the offspring chromosome population to the parent chromosome population, and perform multiple rounds of genetic iteration until the number of genetic iterations reaches the maximum number of generations, complete the iteration, and select the chromosome individual with the highest fitness as the optimal chromosome individual from the genetic iteration process. S378. The optimal path solution corresponding to the optimal chromosome individual is used as the optimal path solution for the UAV. The corresponding optimal path solution is calculated for each UAV, and the optimal path solution corresponding to each UAV is used as the area access path for each UAV.

[0036] In any given round of genetic iteration, among two parental candidate crossover chromosome individuals, firstly, two crossover points are randomly selected, and the segments between these two points are exchanged. Secondly, based on the exchanged segments, a mapping relationship between the genes within the two segments is established. Then, for genes outside the exchanged segments, if they conflict with (i.e., are duplicated) genes in the exchanged segments, they are replaced with non-conflicting genes according to the mapping relationship. Finally, this process is repeated until all genes in the exchanged segments are non-conflicting, thus completing the crossover operation and forming two valid offspring chromosomes.

[0037] In one possible implementation, step S3 involves calculating the corresponding spatial resultant force for each drone based on the drone cluster task configuration data, using the spatial resultant force to generate a corresponding path point sequence for each drone, and forming the access path for each drone based on the path point sequence. This can be decomposed into, but is not limited to, the following steps S38-S19, specifically including: S38. Based on the UAV cluster mission configuration data, construct a three-dimensional simulation mission environment map; S39. In the three-dimensional simulation task environment map, extract the regional center coordinates of each task target area, and mark the regional center coordinates of each task target area as gravity points based on the regional access paths of each UAV. S310. Divide the area access path of each UAV into multiple access path intervals according to each of the gravity points, and set each of the gravity points as the interval endpoint of each access path interval. S311. In the three-dimensional simulation task environment map, calculate the gravitational force exerted on the UAV by the corresponding endpoint of each access path segment; S312. In the three-dimensional simulation task environment map, extract the obstacle information of each task target area, and according to the obstacle information, place each obstacle in each access path interval, so that the obstacle center coordinates of the obstacle in the corresponding access path interval are used as the repulsion point. S313. Calculate the repulsive force generated by the corresponding repulsive point on the UAV in each of the access path intervals; S314. Obtain a preset tangential perturbation gain coefficient. In each segment of the access path interval, when the resultant force of the attraction and the repulsion is 0, generate a corresponding tangential perturbation force for the attraction and the repulsion through the tangential perturbation gain coefficient. S315. Integrate the gravity, the repulsion and the tangential disturbance force to generate the spatial resultant force of each UAV at each moment within each segment of the access path; S316. Extract the drone performance parameters of each drone from the drone cluster task configuration data, and calculate the maximum flight speed of each drone based on the drone performance parameters. S317. Based on the spatial resultant force of each UAV at each moment within each access path interval, in the three-dimensional simulation task environment map, generate the movement direction of each UAV at each moment, and use the movement direction of each UAV at each moment and the maximum flight speed of each UAV to generate the path points of each UAV at each moment. S318. For each UAV, sequentially integrate each path point in each access path interval and the end point of each access path interval to generate a path point sequence corresponding to each UAV. S319. Smooth the path point sequence corresponding to each UAV to obtain the access path of each UAV.

[0038] In a specific application scenario, based on the drone swarm mission configuration data, the corresponding spatial resultant force for each drone is calculated. This spatial resultant force is then used to generate a corresponding path point sequence for each drone. Based on this path point sequence, an access path for each drone is formed. This can be achieved, for example, but not limited to, through the following steps: First, at a specific moment within a certain access path interval, the corresponding gravitational force for the current UAV is calculated using the following formula (4): (4) in, The preset gravitational field gain coefficient, This represents the spatial distance between the current location of the drone (i.e., the path point where the drone is currently located) and the endpoint of the current access path interval. This represents the gravitational pull of the endpoint of the currently accessed path interval on the drone at the current moment; Secondly, at a specific moment within a certain access path interval, the corresponding repulsive force for the current UAV is calculated using the following formula (5): (5) in, These are the indexes of each obstacle in the 3D simulation task environment map. The preset repulsive field gain coefficient, Indicates the current obstacle The radius of the obstacle repulsion field (which can be extracted from the drone swarm mission configuration data). This indicates the current coordinates of the drone and the current obstacles. The spatial distance between the center coordinates of the obstacles (which can be extracted from the drone swarm mission configuration data), Indicates the current obstacle The obstacle radius (which can be extracted from the drone swarm mission configuration data). Indicates the current obstacle at the current moment. Repulsive force on drones; Furthermore, at a certain moment within a specific access path interval, if the resultant force of the gravitational and repulsive forces is 0 at the current moment, the corresponding tangential disturbance force for the current UAV can be calculated using the following formula (6): (6) in, The preset tangential disturbance gain coefficient, This represents the cross product operation. This represents the modulo operation. This represents the tangential disturbance force applied to the current drone at the current moment; Then, at a certain moment within a specific access path interval, based on the current gravity of the drone... repulsive force and tangential disturbance The corresponding spatial resultant force can be calculated using the following formula (7): (7) in, This represents the net spatial force acting on the drone at its current location. Subsequently, for current drones, based on the aforementioned spatial force... The path point at the next moment can be calculated using the following formula (8): (8) in, This indicates the current location (path point) of the drone at the current moment. This indicates the current path point of the drone at the next moment. The current maximum flight speed of the drone (which can be extracted from the drone cluster mission configuration data); Finally, by repeating the calculation process of formulas (4) to (8) at each path point, all path points of the UAV within a certain access path interval can be obtained. By repeating the calculation process for each access path interval, the path point sequence of the UAV is finally formed. Then, by performing path smoothing on the path point sequence of each UAV, the access path of each UAV can be obtained.

[0039] In practical applications, this embodiment successfully overcomes the local minima problem encountered by the traditional artificial potential field method when planning paths, such as in narrow passages or complex obstacle areas, by calculating the spatial resultant force of the UAV at each path point and introducing tangential disturbance force. In multiple (more than 50) simulation tests of UAV swarm scanning tasks in search and rescue scenarios and complex urban scenarios, the UAV swarm collaborative path planning method provided in this embodiment can quickly generate smooth and safe obstacle avoidance paths within 1.5 seconds. This efficient path planning capability can provide accurate, efficient and reliable swarm collaborative paths when dealing with emergency post-disaster search and rescue tasks or complex urban auxiliary tracking tasks.

[0040] S4. Integrate the access paths of each UAV to form a UAV cluster collaborative path planning scheme. Based on the UAV cluster collaborative path planning scheme, issue corresponding access control commands to each UAV to control each UAV to perform access tasks according to the access control commands.

[0041] In one possible implementation, step S4 integrates the access paths of each UAV to form a UAV swarm collaborative path planning scheme. Based on the UAV swarm collaborative path planning scheme, corresponding access control commands are issued to each UAV to control each UAV to perform access tasks according to the access control commands. This can be decomposed into, but is not limited to, the following steps S41-S44, specifically including: S41. Extract drone constraint data from the drone cluster mission configuration data; S42. Based on the task target area allocation table, obtain the access path of each UAV in sequence, and use the UAV constraint data to determine whether there is a collision risk in the access path of each UAV. S43. If not, integrate the access paths of each UAV to form a UAV cluster collaborative path planning scheme. Based on the UAV cluster collaborative path planning scheme, issue corresponding access control instructions to each UAV and add a synchronous take-off trigger instruction to the access control instructions of each UAV to control the UAVs to take off synchronously and execute the access task according to the corresponding access control instructions. S44. If so, then filter out access paths with collision risks, generate corresponding access control instructions for each access path with collision risks, add asynchronous take-off instructions to the access control instructions corresponding to each access path with collision risks, generate corresponding access control instructions for each access path without collision risks, add synchronous take-off trigger instructions to the access control instructions corresponding to each access path without collision risks, so as to control each UAV to take off synchronously / asynchronously and execute the access task according to the corresponding access control instructions.

[0042] It should be noted that the UAV swarm collaborative path planning method provided in this embodiment optimizes the generated collaborative path planning scheme by using UAV constraint data for time control. It not only performs collaborative control of the UAV swarm at the path level, but also performs synchronous / asynchronous control of the working mode and collaborative take-off of the UAV swarm at the time level (without considering collisions and changing the path planning scheme). This allows for fine-tuning of the take-off time of each path with collision risk while ensuring optimal path, forming asynchronous control logic. This significantly improves the safety of collaborative work of the UAV swarm while retaining the excellence of the UAV swarm path planning scheme.

[0043] Furthermore, during control execution, to address the issue of unpredictable external disturbances to UAVs in complex actual working environments, the performance parameters of each UAV can be monitored and updated in real time during flight. This ensures that when each UAV accesses the target area according to the access control instructions, the updated performance parameters can be used to supplement the access control instructions, ensuring that the final execution path is more in line with actual needs. This guarantees the strong adaptability of the UAV swarm to complex external environments and greatly improves the actual task completion rate.

[0044] like Figure 2 As shown, the second aspect of this embodiment provides a hardware system for implementing the UAV swarm cooperative path planning method described in the first aspect of the embodiment, including: The data acquisition unit is used to acquire UAV mission environment data and UAV performance parameter data, and to preprocess the UAV mission environment data and the UAV performance parameter data and calculate secondary parameter data to obtain UAV cluster mission configuration data. The UAV cluster mission configuration data includes multiple mission target areas and UAV performance parameters of each UAV in the UAV cluster. The region allocation unit is used to calculate the corresponding regional spatiotemporal cost for each task target region based on the UAV cluster task configuration data, and generate a task target region allocation table based on the regional spatiotemporal cost of each task target region and the UAV performance parameters of each UAV. The task target region allocation table is used to represent the allocation result of the task target region corresponding to each UAV, and each UAV is allocated at least one task target region. The path generation unit is used to optimize the path for each UAV based on the task target area allocation table, with the goal of minimizing the access distance of a single UAV and the access time of the UAV cluster. It calculates the area access path for each UAV and calculates the corresponding spatial force for each UAV based on the UAV cluster task configuration data. It then uses the spatial force to generate a corresponding path point sequence for each UAV and forms the access path for each UAV based on the path point sequence. The instruction issuing unit is used to integrate the access paths of each UAV to form a UAV cluster collaborative path planning scheme. Based on the UAV cluster collaborative path planning scheme, it issues corresponding access control instructions to each UAV to control each UAV to perform access tasks according to the access control instructions.

[0045] The working process, working details and technical effects of the system provided in this embodiment can be found in the first aspect of the embodiment, and will not be repeated here.

[0046] like Figure 3 As shown, the third aspect of this embodiment provides an electronic device, including: a memory, a processor, and a transceiver that are sequentially and communicatively connected, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the UAV swarm cooperative path planning method as described in the first aspect of the embodiment.

[0047] For specific examples, the memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out (FIFO) memory, and / or first-in-last-out (FILO) memory, etc.; specifically, the processor may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor may be implemented using at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), PLA (Programmable Logic Array). The processor may also include a main processor and a coprocessor. The main processor, also known as the CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state.

[0048] In some embodiments, the processor may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. For example, the processor may not be limited to microprocessors of the STM32F105 series, reduced instruction set computer (RISC) microprocessors, x86 architecture processors, or processors with integrated neural network processing units (NPUs). The transceiver may be, but is not limited to, a Wi-Fi transceiver, a Bluetooth transceiver, a General Packet Radio Service (GPRS) transceiver, a ZigBee transceiver (a low-power LAN protocol based on the IEEE 802.15.4 standard), a 3G transceiver, a 4G transceiver, and / or a 5G transceiver. Furthermore, the device may also include, but is not limited to, a power module, a display screen, and other necessary components.

[0049] The working process, working details and technical effects of the electronic device provided in this embodiment can be found in the first aspect of the embodiment, and will not be repeated here.

[0050] The fourth aspect of this embodiment provides a storage medium for storing instructions containing the UAV swarm cooperative path planning method described in the first aspect of the embodiment. That is, the storage medium stores instructions that, when executed on a computer, perform the UAV swarm cooperative path planning method as described in the first aspect of the embodiment.

[0051] The storage medium refers to a carrier for storing data, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or memory sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0052] The working process, working details and technical effects of the storage medium provided in this embodiment can be found in the first aspect of the embodiment, and will not be repeated here.

[0053] The fifth aspect of this embodiment provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the UAV swarm cooperative path planning method as described in the first aspect of this embodiment, wherein the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0054] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for collaborative path planning in a drone swarm, characterized in that, include: The system acquires UAV mission environment data and UAV performance parameter data, and preprocesses and calculates secondary parameter data on the UAV mission environment data and UAV performance parameter data to obtain UAV cluster mission configuration data. The UAV cluster mission configuration data includes multiple mission target areas and UAV performance parameters of each UAV in the UAV cluster. Based on the UAV cluster task configuration data, the corresponding regional spatiotemporal cost is calculated for each task target area. Based on the regional spatiotemporal cost of each task target area and the UAV performance parameters of each UAV, a task target area allocation table is generated. The task target area allocation table is used to represent the allocation result of the task target area corresponding to each UAV, and each UAV is allocated at least one task target area. With the goal of minimizing the access distance of a single UAV and the access time of a UAV cluster, a path optimization objective function is established for each UAV based on the task target area allocation table to calculate the area access path for each UAV. Based on the UAV cluster task configuration data, the corresponding spatial force is calculated for each UAV to generate a corresponding path point sequence for each UAV using the spatial force. The access path for each UAV is then formed based on the path point sequence. The access paths of each drone are integrated to form a drone cluster collaborative path planning scheme. Based on the drone cluster collaborative path planning scheme, corresponding access control commands are issued to each drone to control each drone to perform access tasks according to the access control commands.

2. The UAV swarm cooperative path planning method according to claim 1, characterized in that, Acquire UAV mission environment data and UAV performance parameter data, and preprocess the UAV mission environment data and UAV performance parameter data to obtain UAV swarm mission configuration data, including: The drone management platform obtains the drone cluster task description, extracts the geographic area information to be accessed from the drone cluster task description, generates multiple task target areas based on the geographic area information to be accessed, and obtains the regional center coordinates, regional boundary vertex coordinates, regional area, regional geometric shape description and regional terrain elevation data of each task target area through the drone management platform to form the task target area information of each task target area. The takeoff base coordinates and return base coordinates are extracted from the drone swarm mission description, and the takeoff base coordinates are used as the starting point of the path planning and the return base coordinates are used as the ending point of the path planning to form initial path planning information; According to the description of the UAV cluster mission, multiple environmental obstacles are extracted from each of the mission target areas, and the obstacle type, obstacle center coordinates and maximum influence radius of each environmental obstacle are obtained through the UAV management platform to form obstacle information for each environmental obstacle; The mission target area information, the initial path planning information, and the obstacle information of each environmental obstacle are integrated into UAV mission environment data. By using the drone management platform, the flight capability parameters, endurance parameters, mission payload parameters, and kinematic parameters of each drone in the drone cluster are obtained to form the drone performance parameters of each drone. By using the drone cluster task description, drone task constraints and drone safety constraints are obtained to form drone constraint data. Obtain a preset task configuration data format, and perform data formatting processing on the UAV task environment data, the UAV performance parameter data and the UAV constraint data through the task configuration data format. Perform coordinate unification processing on the UAV task environment data and the UAV performance parameter data to obtain preprocessed UAV task environment data and preprocessed UAV performance parameter data. Based on the UAV mission environment data and the UAV performance parameter data, the regional distance between each mission target area, the flight time of each UAV between each mission target area, and the scanning time of each UAV to each mission target area are calculated to form secondary parameter data; The preprocessed UAV mission environment data, the preprocessed UAV performance parameter data, and the secondary parameter data are integrated to form UAV swarm mission configuration data.

3. The UAV swarm cooperative path planning method according to claim 1, characterized in that, Based on the aforementioned UAV swarm mission configuration data, the corresponding regional spatiotemporal costs for each mission target area are calculated, including: The spatial neighborhood threshold and the temporal neighborhood threshold are calculated using the drone cluster task configuration data, and based on the spatial neighborhood threshold and the temporal neighborhood threshold, the corresponding spatiotemporal neighborhood task area is selected for each of the task target areas. For each of the task target regions, the scanning time cost between each of the task target regions and the corresponding spatiotemporal neighbor task regions is calculated using the following formula (1): （1） in, and All of these are indexes of the target region of the task. The total number of the target areas for the task. It is a binary function. Indicates the target area of ​​the task With the target area of ​​the task The distance between the regions, Indicates that the drone is in the mission target area With the target area of ​​the task Flight time between Indicates that the drone is in the mission target area Scanning time within, Indicates that the drone is in the mission target area Scanning time within, This represents the spatial neighborhood threshold. This represents the time neighborhood threshold. Indicates the target area of ​​the task The scanning time cost between the corresponding spatiotemporal neighboring task regions; The corresponding scanning time cost is calculated for each of the mission target areas, and the flight space cost corresponding to each of the mission target areas is calculated based on the scanning time cost corresponding to each of the mission target areas. Based on the scanning time cost and flight space cost of each UAV in each of the aforementioned mission target areas, the regional spatiotemporal cost corresponding to each mission target area is calculated using the following formula (2): （2） in, Indicates the target area of ​​the mission The cost of flight space, Indicates the target area of ​​the mission Regional time and space costs.

4. The UAV swarm cooperative path planning method according to claim 3, characterized in that, Based on the spatiotemporal cost of each mission objective region and the performance parameters of each UAV, a mission objective region allocation table is generated, including: Based on the regional spatiotemporal cost of each of the task target regions, the task target regions are arranged in descending order of regional spatiotemporal cost to form a task target region sequence. Based on the drone swarm mission configuration data, performance coefficients are generated for each drone, and the drones are arranged in descending order of performance coefficients to form a drone swarm sequence. Based on the mission target area sequence and the UAV cluster sequence, a corresponding area allocation relationship is established between each UAV and each mission target area. The area allocation relationships corresponding to each UAV are integrated to generate a mission target area allocation table.

5. The UAV swarm cooperative path planning method according to claim 4, characterized in that, Based on the mission target region sequence and the UAV cluster sequence, a corresponding region allocation relationship is established between each UAV and each mission target region. The region allocation relationships corresponding to each UAV are integrated to generate a mission target region allocation table, including: The number of drones in the drone swarm sequence and the number of mission target areas in the mission target area sequence are compared to obtain the comparison results. When the comparison result indicates that the number of drones in the drone swarm sequence is greater than the number of mission target areas in the mission target area sequence, the method includes: The last drone in the drone cluster sequence is removed sequentially until the number of drones in the drone cluster sequence is equal to the number of task target areas in the task target area sequence. Each task target region in the task target region sequence is assigned one-to-one to each drone in the drone cluster sequence after the removal is completed, forming a region allocation relationship for each drone. The region allocation relationships for each drone are then integrated to generate a task target region allocation table. When the comparison result indicates that the number of drones in the drone swarm sequence is no more than the number of mission target areas in the mission target area sequence, the method includes: According to the mission target area sequence, the mission target area is allocated to each drone in the drone cluster sequence to form a core scanning area corresponding to each drone. The core scanning area corresponding to each drone is used as the first element to establish a scanning area set for each drone. According to the mission target area sequence, the scanning area sets corresponding to each UAV are arranged in order to form the UAV scanning area sequence, and according to the mission target area sequence, the unassigned mission target areas are arranged in reverse order to form the unassigned area sequence. The unallocated region sequence is filled into the scan region set of the UAV scan region sequence in the corresponding order, sorted after the first element to form a pre-filled scan region set, and the total access time of the pre-filled scan region set is calculated, wherein the total access time includes the scanning time of the UAV scanning each task target region in the pre-filled scan region set and the shortest time of the UAV flying between each task target region in the pre-filled scan region set. The flight endurance parameters of each drone are extracted from the drone cluster task configuration data. The total flight endurance of each drone is calculated based on the flight endurance parameters of each drone. The total flight endurance of each drone is used to verify the total access time of the pre-filled scan area set corresponding to each drone, and the verification results are obtained. If the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone does not exceed the total flight time of the current drone, then this filling is performed to update the pre-filled scan area set to a scan area set. The unallocated areas in the unallocated area sequence are then used to fill the scan area set in order to form a pre-filled scan area set again. The verifiable verification is then performed again until the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone exceeds the total flight time of the current drone. If the verification result is that the total access time of the pre-filled scan area set corresponding to the current drone exceeds the total flight time of the current drone, then the filling will not be performed, the scan area set corresponding to the current drone will be used as the final scan area set, and the scan area set corresponding to the current drone will be removed from the drone scan area sequence to complete the update of the drone scan area sequence. The task target regions in the unallocated region sequence are sequentially filled into the UAV scanning region sequence until the UAV scanning region sequence is empty. The final scanning region set corresponding to each UAV is obtained. The final scanning region set corresponding to each UAV is used as the region allocation relationship corresponding to each UAV. The region allocation relationships corresponding to each UAV are integrated to generate a task target region allocation table.

6. The UAV swarm cooperative path planning method according to claim 1, characterized in that, With the objectives of minimizing the individual drone access distance and the drone swarm access time, a path optimization objective function is established for each drone based on the task objective area allocation table to calculate the area access path for each drone, including: Extract the path planning start point and path planning end point from the drone cluster mission configuration data; Based on the task target area allocation table, the path planning start point, and the path planning end point, a corresponding access distance minimization solution function is established for each UAV. The access distance is the sum of the flight distance between the UAV from the path planning start point to the corresponding initial task target area in the task target area allocation table, the flight distance between each task target area in the task target area allocation table, and the flight distance between the UAV from the corresponding final task target area in the task target area allocation table to the path planning end point. Based on the task target area allocation table, the path planning start point, and the path planning end point, a corresponding access time maximization function is established for each UAV. The access time is the sum of the UAV's flight time from the path planning start point to the corresponding initial task target area in the task target area allocation table, the UAV's flight time between each task target area in the task target area allocation table, the UAV's scan time within each task target area in the task target area allocation table, and the UAV's flight time from the corresponding final task target area in the task target area allocation table to the path planning end point. The function for minimizing the access distance of each drone is used as the optimization function for the shortest access distance of each individual drone. The minimum value of the function for maximizing the access time of each drone is obtained to get the optimization function for the shortest access time of the drone cluster. Obtain a preset target weight, and sum the access distance minimization solution function and the drone cluster access time shortest optimization function based on the target weight to form a path optimization objective function for each drone. For each UAV, an initial path solution is generated based on the path optimization objective function, so as to form a path solution space for each UAV based on each path solution. Based on the genetic algorithm, the path solution space corresponding to each UAV is optimized to obtain the optimal path solution for each UAV, and the optimal path solution is used as the area access path for each UAV.

7. The UAV swarm cooperative path planning method according to claim 6, characterized in that, Based on a genetic algorithm, the path solution space corresponding to each UAV is optimized to obtain the optimal path solution for each UAV. This optimal path solution is then used as the area access path for each UAV, including: In the path solution space corresponding to each UAV, a path solution is taken as a chromosome individual to form an initial chromosome population. The path solutions corresponding to the order of the task target region in the task target region allocation table are taken as initial elite chromosome individuals and added to the initial chromosome population. The initial chromosome population is taken as the parent chromosome population. Based on the function value of the path optimization objective function corresponding to each individual chromosome, the fitness of each individual chromosome in the parent chromosome population is calculated using the following formula (3): （3） in, This is the index of the individual chromosome. Individuals with chromosomes The corresponding function value of the path optimization objective function, For division by zero parameters, Individuals with chromosomes Corresponding fitness; A preset elite individual retention ratio is obtained. Based on the elite individual retention ratio and the number of chromosome individuals in the parent chromosome population, the number of chromosomes to be retained is calculated. The multiple chromosome individuals with the highest fitness are selected as elite chromosome individuals and inherited to the offspring chromosome population. The number of elite chromosome individuals is determined by the number of chromosomes to be retained. A preset parent candidate ratio is obtained. Based on the parent candidate ratio and the number of chromosome individuals in the parent chromosome population, the number of parent candidates is calculated. Using a roulette wheel method, multiple chromosome individuals are selected from the initial chromosome population as parent candidate crossover chromosome individuals. The number of parent candidate crossover chromosome individuals is determined by the number of parent candidates, and the sum of the parent candidate ratio and the elite individual retention ratio is 1. Obtain a preset crossover inheritance probability, perform partial mapping crossover on each of the parent candidate crossover chromosome individuals based on the crossover inheritance probability to obtain crossover offspring chromosomes, and supplement the crossover offspring chromosomes to the pre-offspring chromosome population; A preset mutation probability is obtained, and a mutation operation is performed on the offspring chromosome population based on the mutation probability to obtain an offspring chromosome population. The mutation operation includes exchange mutation, reverse mutation, and insertion mutation. Obtain the preset maximum number of generations, update the offspring chromosome population to the parent chromosome population, and perform multiple rounds of genetic iteration until the number of genetic iterations reaches the maximum number of generations, complete the iteration, and select the chromosome individual with the highest fitness as the optimal chromosome individual from the genetic iteration process. The optimal path solution corresponding to the optimal chromosome individual is used as the optimal path solution for the UAV. The corresponding optimal path solution is calculated for each UAV, and the optimal path solution corresponding to each UAV is used as the area access path for each UAV.

8. The UAV swarm cooperative path planning method according to claim 6, characterized in that, Based on the drone swarm mission configuration data, the corresponding spatial resultant force is calculated for each drone. This spatial resultant force is then used to generate a corresponding path point sequence for each drone. Based on this path point sequence, an access path for each drone is formed, including: Based on the aforementioned UAV cluster mission configuration data, a three-dimensional simulation mission environment map is constructed. In the three-dimensional simulation mission environment map, the regional center coordinates of each mission target area are extracted, and based on the regional access paths of each UAV, the regional center coordinates of each mission target area are marked as gravity points. The area access path of each UAV is divided into multiple access path intervals according to each gravity point, and each gravity point is set as the interval endpoint of each access path interval. In the three-dimensional simulation mission environment map, the gravitational force exerted on the UAV by the corresponding endpoint of each access path segment is calculated. In the three-dimensional simulation task environment map, obstacle information of each task target area is extracted, and according to the obstacle information, each obstacle is placed in each access path interval, so that the obstacle center coordinates of the obstacle in the corresponding access path interval are used as the repulsion point. In each of the access path intervals, the repulsive force generated by the corresponding repulsive point on the drone is calculated; Obtain a preset tangential perturbation gain coefficient. In each segment of the access path interval, when the resultant force of the attraction and the repulsion is 0, generate a corresponding tangential perturbation force for the attraction and the repulsion through the tangential perturbation gain coefficient. By integrating the gravitational force, the repulsive force, and the tangential perturbation force, the spatial resultant force of each UAV at each moment within each segment of the access path is generated; The performance parameters of each drone are extracted from the drone swarm mission configuration data, and the maximum flight speed of each drone is calculated based on these parameters. Based on the spatial resultant force of each UAV at each moment within each access path interval, the movement direction of each UAV at each moment is generated in the three-dimensional simulation task environment map. Using the movement direction of each UAV at each moment and the maximum flight speed of each UAV, the path points of each UAV at each moment are generated. For each drone, the path points in each access path interval and the end point of each access path interval are sequentially integrated to generate a path point sequence corresponding to each drone. The path point sequence corresponding to each UAV is smoothed to obtain the access path of each UAV.

9. The UAV swarm cooperative path planning method according to claim 1, characterized in that, The access paths of each drone are integrated to form a drone swarm collaborative path planning scheme. Based on the drone swarm collaborative path planning scheme, corresponding access control commands are issued to each drone to control each drone to perform access tasks according to the access control commands, including: Extract drone constraint data from the drone cluster mission configuration data; Based on the task target area allocation table, the access paths of each UAV are obtained sequentially, and the UAV constraint data is used to determine whether there is a collision risk in the access paths of each UAV. If not, the access paths of each UAV are integrated to form a UAV cluster collaborative path planning scheme. Based on the UAV cluster collaborative path planning scheme, corresponding access control commands are issued to each UAV, and a synchronous take-off trigger command is added to the access control command corresponding to each UAV to control each UAV to take off synchronously and execute the access task according to the corresponding access control command. If so, then the access paths with collision risks are selected, and corresponding access control instructions are generated for each access path with collision risks. An asynchronous take-off instruction is added to the access control instructions corresponding to each access path with collision risks. For each access path without collision risks, corresponding access control instructions are generated, and a synchronous take-off trigger instruction is added to the access control instructions corresponding to each access path without collision risks, so as to control each UAV to take off synchronously / asynchronously and execute the access task according to the corresponding access control instructions.

10. A collaborative path planning system for unmanned aerial vehicle (UAV) swarms, characterized in that, The method for collaborative path planning of unmanned aerial vehicle (UAV) swarms as described in any one of claims 1 to 9 includes: The data acquisition unit is used to acquire UAV mission environment data and UAV performance parameter data, and to preprocess the UAV mission environment data and the UAV performance parameter data and calculate secondary parameter data to obtain UAV cluster mission configuration data. The UAV cluster mission configuration data includes multiple mission target areas and UAV performance parameters of each UAV in the UAV cluster. The region allocation unit is used to calculate the corresponding regional spatiotemporal cost for each task target region based on the UAV cluster task configuration data, and generate a task target region allocation table based on the regional spatiotemporal cost of each task target region and the UAV performance parameters of each UAV. The task target region allocation table is used to represent the allocation result of the task target region corresponding to each UAV, and each UAV is allocated at least one task target region. The path generation unit is used to optimize the path for each UAV based on the task target area allocation table, with the goal of minimizing the access distance of a single UAV and the access time of the UAV cluster. It calculates the area access path for each UAV and calculates the corresponding spatial force for each UAV based on the UAV cluster task configuration data. It then uses the spatial force to generate a corresponding path point sequence for each UAV and forms the access path for each UAV based on the path point sequence. The instruction issuing unit is used to integrate the access paths of each UAV to form a UAV cluster collaborative path planning scheme. Based on the UAV cluster collaborative path planning scheme, it issues corresponding access control instructions to each UAV to control each UAV to perform access tasks according to the access control instructions.

Images