Unmanned aerial vehicle cluster stealth flight path planning optimization design method based on task allocation

Abstract

The invention discloses an unmanned aerial vehicle cluster stealth flight path planning optimization design method based on task allocation, and the method comprises the steps: designing an unmanned aerial vehicle cluster stealth flight path planning comprehensive cost function based on task allocation based on the voyage cost of an unmanned aerial vehicle flight path, the degree of networking radar threat and the degree of antiaircraft gun threat; taking minimization of an unmanned aerial vehicle cluster stealth flight path planning comprehensive cost function based on task allocation as an optimization target, and taking meeting of flight path feasibility judgment, unmanned aerial vehicle cluster task allocation scheme requirements and unmanned aerial vehicle cluster dynamics limitation as constraint conditions, constructing an unmanned aerial vehicle cluster stealth flight path planning optimization model based on task allocation; and solving the optimization model by adopting an improved algorithm based on a genetic algorithm and an improved A * algorithm to obtain an optimal task allocation scheme of the unmanned aerial vehicle cluster. The application environment is more complex, the stealth performance of the unmanned aerial vehicle cluster is effectively improved, and meanwhile the comprehensive cost of task execution of the unmanned aerial vehicle cluster is effectively reduced.

Description

Technical Field

[0001] The present invention relates to radar signal processing technology, and in particular to an optimization design method for stealth track planning of unmanned aerial vehicle clusters based on task allocation. Background Art

[0002] With the continuous development of the drone field, its application scenarios are constantly expanding, and its value is constantly increasing. However, when dealing with a diverse environment, a single drone faces many limitations and it is difficult to meet application requirements. Therefore, the advantages of drone swarm collaboration are obvious. Drone swarm trajectory planning refers to planning a trajectory that meets certain indicators from the starting point to the end point for the drone swarm under the constraints of drone swarm performance, penetration safety, threat environment, etc. Drone swarm trajectory planning is the basis for drone swarms to perform various complex tasks, and can effectively improve the performance of drone swarms in performing complex tasks. Effective task allocation for drone swarms can significantly improve the effectiveness and survivability of drone mission execution and reduce the resource consumption of drone swarms.

[0003] At present, there are many studies on UAV task allocation and trajectory planning. However, most of the studies do not consider the radar cross section (RCS) characteristics of the UAV itself, and most of them are aimed at a single UAV, which is difficult to meet the needs of the actual environment. In summary, there is no optimization design method for UAV cluster stealth trajectory planning based on task allocation in the existing technology. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a method for optimizing the design of stealth trajectory planning for drone clusters based on task allocation, which is applicable to more complex environments and can effectively improve the stealth performance of drone clusters while reducing the resource consumption of drone clusters.

[0005] Technical solution: The method for optimizing the design of stealth trajectory planning of a UAV cluster based on task allocation of the present invention comprises the following steps:

[0006] Based on the range cost of the UAV trajectory, the threat level of the networked radar and the threat level of the anti-aircraft gun, a comprehensive cost function for the stealth trajectory planning of the UAV swarm based on task allocation is designed;

[0007] Taking minimizing the comprehensive cost function of UAV cluster stealth trajectory planning based on task allocation as the optimization goal, and satisfying the trajectory feasibility judgment, UAV cluster task allocation plan requirements and UAV cluster dynamics constraints as constraints, an optimization model for UAV cluster stealth trajectory planning based on task allocation is constructed.

[0008] An improved algorithm based on genetic algorithm and improved A* algorithm is used to solve the optimization model and obtain the optimal task allocation plan for the UAV cluster.

[0009] Furthermore, the expression of the comprehensive cost function of the UAV cluster stealth trajectory planning based on task allocation is:

[0010]

[0011] in, is the comprehensive cost function of stealth trajectory planning of UAV swarm based on task allocation, M is the number of UAVs in the UAV swarm; represents the flight cost of the mth UAV track, which is evaluated by the UAV flight range, K m is the number of track segments contained in the track of the mth UAV, l m,k is the distance of the kth track segment of the mth UAV; is the threat level of the networked radar to the track of the mth UAV, which is evaluated by the probability and value of radar detection received by the track point; T m is the flight time of the mth drone; is the probability of the mth UAV being detected by the networked radar at time t; is the degree of anti-aircraft gun threat to the track of the mth UAV, which is evaluated by the probability and probability of anti-aircraft gun threat to the track point; is the probability that the mth UAV is threatened by anti-aircraft guns at time t; ω1, ω2, and ω3 are the weights of the range, the threat level of the networked radar, and the threat level of the anti-aircraft guns, respectively.

[0012] Furthermore, the expression of the optimization model of UAV cluster stealth trajectory planning based on task allocation is:

[0013]

[0014] in, is the comprehensive cost function of UAV cluster stealth trajectory planning based on task allocation, M is the number of UAVs in the UAV cluster, K m is the number of track segments contained in the track of the mth UAV, l m,k is the distance of the kth track segment of the mth UAV, T m is the flight time of the mth drone, is the probability of the mth UAV being detected by the networked radar at time t; is the probability that the mth UAV is threatened by anti-aircraft guns at time t; ω1, ω2, and ω3 are the weights of the range, the threat level of the networked radar, and the threat level of the anti-aircraft guns, respectively; L max is the maximum range of the drone, is the heading angle of the mth UAV at time t, is the heading angle of the mth UAV at time t+1, △ψ max is the maximum turning angle of each UAV; P c is the detection probability threshold of the networked radar, P a is the probability threshold of anti-aircraft gun detection, is the spatial position of the mth UAV at time t, obstacle is the spatial range of the fixed obstacle, σ is the RCS of the UAV according to the angle distribution, and σ q is the RCS of the drone at the qth angle, is the spatial position of the jth UAV at time t, d safe is the safety distance, x mn is the UAV cluster task allocation scheme, m is the UAV number, n is the task number, and N is the total number of tasks in the UAV cluster.

[0015] Furthermore, an improved algorithm based on genetic algorithm and improved A* algorithm is used to solve the optimization model, including: first, the improved A* algorithm is used to calculate the comprehensive track cost of each UAV performing different tasks, and the comprehensive cost and track corresponding to each UAV performing different tasks are stored in table S; then, based on table S, the genetic algorithm is used to solve the optimal task allocation plan for the UAV cluster.

[0016] Furthermore, the heuristic function expression used in the improved A* algorithm is:

[0017] f(n0)=g(n0)+ωh(n0)

[0018] Among them, f(n0) is the comprehensive cost function from the starting point to the end point through the track point n0; g(n0) is the actual cost function from the starting point to the track point n0; h(n0) is the estimated cost function from the track point n0 to the end point of the track, represented by the Manhattan distance; ω is the weighting factor.

[0019] Furthermore, the actual cost function g(n0) of the drone from the starting point to the track point n0 is expressed as:

[0020]

[0021] in, The actual flight distance of the mth UAV from the starting point to the track point n0, l m,k is the distance of the kth track segment of the mth UAV, is the number of track segments from the starting point to the track point n0 of the mth UAV, is the time taken by the mth UAV to travel from the starting point to the track point n0, It represents the probability of network radar detection from the starting point to the track point n0, is the probability that the mth UAV is detected by the networked radar at time t, represents the probability and threat probability of anti-aircraft guns on the track from the starting point to the track point n0. is the probability that the mth UAV is threatened by anti-aircraft guns at time t, ω1, ω2, and ω3 are the weights of the range, the threat level of the networked radar, and the threat level of the anti-aircraft guns, respectively.

[0022] Furthermore, the fitness function in the genetic algorithm is expressed as the inverse of the comprehensive cost of the cluster under different task allocation schemes.

[0023] The system corresponding to the method includes:

[0024] The comprehensive cost function design unit is used to design a comprehensive cost function for the stealth trajectory planning of UAV clusters based on task allocation, based on the range cost of the UAV trajectory, the threat level of the networked radar and the threat level of the anti-aircraft gun;

[0025] The optimization model building unit is used to take minimizing the comprehensive cost function of the UAV cluster stealth trajectory planning based on task allocation as the optimization goal, and to meet the trajectory feasibility determination, the requirements of the UAV cluster task allocation plan and the UAV cluster dynamics constraints as the constraints, to build an optimization model for the UAV cluster stealth trajectory planning based on task allocation;

[0026] The optimization model solving unit is used to solve the optimization model by adopting an improved algorithm based on genetic algorithm and improved A* algorithm to obtain the optimal task allocation plan for the drone cluster.

[0027] An electronic device for storing and executing the method, the device comprising:

[0028] A memory storing executable program code;

[0029] a processor coupled to the memory;

[0030] The processor calls the executable program code stored in the memory to execute the steps of the method for optimizing the design of stealth trajectory planning of a swarm of unmanned aerial vehicles based on task allocation.

[0031] A computer-readable storage medium for storing and executing the method, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the steps of the method for optimizing the design of stealth trajectory planning for a swarm of unmanned aerial vehicles based on task allocation.

[0032] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are: by optimizing the design of parameters such as the flight heading angle of the task allocation scheme of the drone cluster and the trajectory, the comprehensive cost function of the drone cluster stealth trajectory planning based on task allocation is minimized under the constraints of the trajectory feasibility judgment, the requirements of the drone cluster task allocation scheme and the dynamic limitations of the drone cluster; the applicable scenarios are more complex, and the optimal cluster trajectory design based on task allocation is realized, which effectively improves the stealth performance of the drone cluster, and at the same time effectively reduces the comprehensive cost of the drone cluster to perform tasks, that is, reduces the resource consumption of the drone cluster. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a flow chart of the method of the present invention;

[0034] Figure 2 Schematic diagram of the drone's extended search process;

[0035] Figure 3 Schematic diagram of drone RCS distribution;

[0036] Figure 4 UAV swarm task allocation results and trajectory diagram;

[0037] Figure 5 The angle change diagram of the drone cluster’s track;

[0038] Figure 6 Probability change diagram of drone cluster track;

[0039] Figure 7 Comparison of algorithmic drone cluster trajectory probability changes. DETAILED DESCRIPTION

[0040] The structure and working process of the present invention are further described below in conjunction with the accompanying drawings and specific embodiments.

[0041] The method for optimizing and designing the stealth trajectory planning of a swarm of unmanned aerial vehicles based on task allocation described in the present invention optimizes the trajectory of a swarm of unmanned aerial vehicles under the constraints of satisfying the feasibility of the trajectory, the requirements of the task allocation scheme of the swarm of unmanned aerial vehicles, and the dynamic limitations of the swarm of unmanned aerial vehicles, thereby improving the stealth performance of the swarm of unmanned aerial vehicles and reducing the resource consumption of the swarm of unmanned aerial vehicles. Firstly, in combination with the RCS of the unmanned aerial vehicles themselves, a comprehensive cost function for stealth trajectory planning of swarm of unmanned aerial vehicles based on task allocation is designed; secondly, taking minimizing the comprehensive cost function of the cluster as the optimization goal, a stealth trajectory planning optimization model for swarm of unmanned aerial vehicles based on task allocation is constructed, and the task allocation scheme of the swarm of unmanned aerial vehicles and the flight heading angle of the trajectory and other parameters are optimized to achieve the purpose of improving the stealth performance of the swarm trajectory and reducing the resource consumption of the swarm of unmanned aerial vehicles. Figure 1 As shown, the following steps are included:

[0042] S1. Design a comprehensive cost function for UAV cluster stealth trajectory planning based on task allocation;

[0043] The UAV cluster task allocation scheme can be expressed as:

[0044]

[0045] Among them, x mn is the UAV cluster task allocation scheme, m is the UAV number, n is the task number, N is the total number of tasks in the drone cluster, and M is the number of drones in the drone cluster.

[0046] When x mn Corresponding to different values, there are different drone cluster task allocation schemes. Under different task allocation schemes, the drone cluster will plan different trajectories.

[0047] Therefore, the comprehensive cost function of the UAV cluster stealth trajectory planning based on task allocation designed by the present invention is:

[0048]

[0049] in, is the comprehensive cost function of stealth trajectory planning of UAV swarm based on task allocation, M is the number of UAVs in the UAV swarm; represents the flight cost of the mth UAV track, which is evaluated by the UAV flight range, K m is the number of track segments contained in the track of the mth UAV, l m,k is the distance of the kth track segment of the mth UAV; is the threat level of the networked radar to the track of the mth UAV, which is evaluated by the probability and value of radar detection received by the track point; T m is the flight time of the mth drone; is the probability of the mth UAV being detected by the networked radar at time t; is the degree of anti-aircraft gun threat to the track of the mth UAV, which is evaluated by the probability and probability of anti-aircraft gun threat to the track point; is the probability that the mth UAV is threatened by anti-aircraft guns at time t; ω1, ω2, ω3 are the weights of the range, the threat level of the networked radar, and the threat level of the anti-aircraft guns, respectively, and ω1+ω2+ω3=1, where the value ranges of ω1, ω2, and ω3 are all [0,1].

[0050] S2. Construct a stealth trajectory planning optimization model for UAV swarm based on task allocation;

[0051] Assume that each drone has the same performance. Since the drone carries limited fuel, the flight range of each drone should be less than the maximum range L max ,Right now:

[0052]

[0053] When the drone changes its heading, it will be limited by the turning angle due to inertia. The turning angle of the drone should meet the following requirements:

[0054]

[0055] in, is the heading angle of the mth UAV at time t, that is, the angle between the velocity direction of the mth UAV at time t and the positive direction of the right-hand rectangular coordinate X axis; is the heading angle of the mth UAV at time t+1; △ψ max is the maximum turning angle of the mth UAV.

[0056] To ensure the successful planning of the UAV cluster trajectory, the UAVs must fly around fixed obstacles, namely:

[0057]

[0058] in, is the spatial position of the mth UAV at time t; obstacle is the spatial range of the fixed obstacle.

[0059] In addition, the collision avoidance problem of each drone in the drone cluster must be considered, which can be expressed as:

[0060]

[0061] in, is the spatial position of the jth UAV at time t; d safe For a safe distance.

[0062] Therefore, with minimizing the comprehensive cost function of the UAV cluster stealth trajectory planning based on task allocation as the optimization goal, and satisfying the trajectory feasibility judgment, the requirements of the UAV cluster task allocation plan and the UAV cluster dynamics constraints as the constraints, an optimization model of the UAV cluster stealth trajectory planning based on task allocation is constructed, as shown in the following formula:

[0063]

[0064]

[0065] Among them, σ is the RCS of the drone according to the angle distribution, σ q is the RCS of the drone at the qth angle; is the probability of the mth UAV being detected by the network radar at time t, which must be less than a certain threshold P to ensure the safety of the platform c , P c is the detection probability threshold of the networked radar; is the probability of the mth UAV being detected by the anti-aircraft gun at time t, which must be less than a certain threshold P to ensure the safety of the platform a , P a is the probability threshold of anti-aircraft gun detection.

[0066] S3, using an improved algorithm based on genetic algorithm and improved A* algorithm to solve the optimization model;

[0067] In order to solve the optimization model (7), the present invention proposes an improved algorithm based on a genetic algorithm and an improved A* algorithm. The specific steps are as follows:

[0068] Step 1: First, use the improved A* algorithm to calculate the comprehensive cost of stealth trajectory planning for each UAV to perform different tasks. Take the mth UAV performing task n as an example.

[0069] The heuristic function used in the improved A* algorithm is as follows:

[0070] f(n0)=g(n0)+ωh(n0) (8)

[0071] Among them, f(n0) is the comprehensive cost function from the starting point to the end point through the track point n0; g(n0) is the actual cost function from the starting point to the track point n0; h(n0) is the estimated cost function from the track point n0 to the end point of the track, represented by the Manhattan distance; ω is the weighting factor.

[0072] The actual cost function of the drone from the starting point to the track point n0 can be expressed as:

[0073]

[0074] in, The actual distance of the mth UAV from the starting point to the track point n0; is the number of track segments from the starting point to the track point n0 of the mth UAV; is the time taken by the mth UAV to travel from the starting point to the track point n0; It represents the probability and probability of network radar detection for the section of track from the starting point to track point n0; It represents the probability and value of anti-aircraft gun threat on the track from the starting point to track point n0.

[0075] Compared with the standard A* algorithm, the heuristic functions of the improved A* algorithm and the standard A* algorithm are both composed of an actual cost function and an estimated cost function, but the difference is that the actual cost function g(n0) of the improved A* algorithm takes into account multiple threats and the drone's own RCS, while the actual cost function of the standard A* algorithm only considers distance.

[0076] The specific method of trajectory optimization design for a single UAV to perform a single mission is as follows:

[0077] 1) The drone starts from the starting point and first calculates the positions of all possible next track points of the current track point;

[0078] 2) The incident angle of the radar transmission wave on the UAV can be calculated based on the position of the next possible track point, and then the corresponding RCS can be obtained based on the UAV RCS distribution map;

[0079] 3) According to the UAV dynamics constraints, determine whether the next possible track point is expandable. If it is expandable, calculate the estimated cost function, and then calculate the comprehensive cost of reaching the end point through the next possible track point according to formula (8);

[0080] 4) All possible next track points that can be expanded are loaded into the OPEN table, and the optimal next track point is selected by comparing the comprehensive cost of each possible track point. This process is continued until the end point, and the optimization is completed. The UAV expansion search process is as follows: Figure 2 As shown, Figure 2 In , A0, A1, A2, and A3 are all track points of the track obtained by the final optimization design. If A0 is the current track point, A1 is the optimal next track point obtained by optimization, A2 is the optimal next track point of A1, and A3 is the optimal next track point of A2. Figure 2 It is intended to illustrate the optimization process of track points during track optimization design (i.e., selection of track points).

[0081] After the mth UAV completes task n, it will re-execute the next task. After completing all tasks, it will start calculating and planning the trajectory of the m+1th UAV to perform all different tasks. When all UAVs have performed all tasks, the comprehensive cost and trajectory corresponding to each UAV performing different tasks are stored in table S.

[0082] Step 2: Use genetic algorithm to solve the optimal task allocation solution for drone clusters;

[0083] Genetic algorithm solves the optimal task allocation plan for drone clusters. First, the task allocation plan needs to be initialized.

[0084] Different task allocation schemes can be represented by arrays. Assuming there are 6 drones in total, a certain task allocation scheme can be represented as [1 2 3 4 5 6], that is, the first drone performs task 1, the second drone performs task 2, ... the sixth drone performs task 6. The following are the specific steps for using genetic algorithms to solve the optimal task allocation scheme for drone clusters:

[0085] 1) Genetic algorithm is used to randomly initialize the population, that is, to generate a series of UAV cluster task allocation schemes Initial;

[0086] 2) Then query table S to obtain the UAV cluster trajectory and comprehensive cost corresponding to each group of UAV cluster task allocation schemes;

[0087] 3) According to the cluster track and comprehensive cost corresponding to the UAV task allocation scheme obtained from the query table S, determine whether the obtained UAV cluster track meets the UAV collision avoidance condition; if it does, the corresponding cluster track and comprehensive cost in Table S are the cluster track and comprehensive cost under this task scheme; if it does not meet the requirements, it is necessary to re-plan the UAV cluster track that meets the safety conditions according to the specific method 1) to step 4) of track optimization design in step 1 under this group of UAV cluster task allocation schemes, and obtain a new cluster track and comprehensive cost;

[0088] 4) The inverse of the cluster comprehensive cost under different task allocation schemes is used as the fitness function of the corresponding task allocation scheme;

[0089] 5) According to the fitness function value, the initial series of task allocation schemes Initial are selected, crossed, mutated, reversed, and re-inserted into offspring to obtain a new population;

[0090] 6) Determine whether it converges or reaches the maximum number of iterations. If so, the optimization ends; if not, return to 2).

[0091] Therefore, the UAV cluster task allocation plan with the minimum cluster comprehensive cost can be obtained, and the corresponding UAV cluster trajectory is the required trajectory.

[0092] Simulation results:

[0093] Assume that the drone cluster consists of six drones with the same performance, the flight area is 500km×500km, and the network radar detection probability threshold P c =0.3. Assuming that the performance of each drone is the same, the speed is constant at 139m / s, and the maximum turning angle does not exceed The search step length is 5km. To facilitate calculation, the heading angle of the drone is uniformly defined as is the angle between the velocity direction of the mth UAV at time t and the positive direction of the right-hand rectangular coordinate X axis, ranging from [0,2π). The starting and ending points of the UAVs are shown in Table 1. The RCS distribution of the UAVs is shown in the attached figure. Figure 3 shown.

[0094] The simulation scene consists of radar, anti-aircraft guns, and fixed obstacles; the radar parameters are shown in Appendix 2; there are two fixed obstacles, and the location range is shown in Appendix 3; the range weight w1 is set to 0.01, the anti-aircraft gun threat level weight w3 is set to 0.1, and the networked radar threat level weight w2 is set to 1-w1-w3; the anti-aircraft gun parameter settings are shown in Appendix 4.

[0095] Table 1 UAV start and end point settings

[0096]

[0097] Table 2 Radar start and end point settings

[0098]

[0099] Table 3 Fixed obstacle range

[0100]

[0101] Table 4 Parameter settings of anti-aircraft gun in scenario 3

[0102]

[0103] UAV swarm task allocation and trajectory simulation results are shown in the figure below. Figure 4 As shown in Figure 1, the task allocation scheme is: UAV1→Task1; UAV2→Task5; UAV3→Task2; UAV4→Task4; UAV5→Task3; UAV6→Task6. Under the optimal task allocation scheme, the heading angle changes of each UAV during the flight of the UAV cluster are as follows: Figure 5 As shown in the figure, during the flight of the drone cluster, the probability of each drone track being detected by the networked radar at each moment changes as shown in the figure. Figure 6 As shown. Figure 4 , Figure 5 , Figure 6 It can be seen from the results that the method of the present invention can successfully optimize the design of the UAV cluster stealth trajectory planning based on task allocation, solve the optimal task allocation plan, and plan the UAV cluster trajectory that meets the dynamic constraints and ensures the safety of the platform, which illustrates the effectiveness of the proposed method.

[0104] In the same scenario, the genetic algorithm is also used to optimize the task allocation scheme, and the standard A* method is used as a comparison method to plan the trajectory of the drone cluster, and the algorithm of the present invention is compared with the comparison algorithm. The calculation results of the radar detection probability of the drone cluster trajectory planned by the comparison algorithm are as follows Figure 7 shown.

[0105] Depend on Figure 7 It can be seen that the comparison algorithm fails to plan the trajectory while allocating tasks. This verifies the superiority of the method of the present invention. Temporarily ignoring the detection probability threshold of the networked radar, the cluster trajectory range and comprehensive cost planned by the method of the present invention and the comparison algorithm are compared, and the comparison results are shown in Table 5. It can be seen from the results in Table 5 that the range cost paid by the method of the present invention is slightly higher, but the comprehensive cost is lower, which verifies the superiority of the method of the present invention.

[0106] Table 5 Comparison of the proposed algorithm and the comparison algorithm in scenario 3

[0107]

[0108] The working principle and working process of the invention are as follows:

[0109] Starting from the actual combat scenario, the present invention considers the two-dimensional multi-threat scenario composed of radar, anti-aircraft guns, and fixed obstacles. The drone cluster combines its own radar scattering characteristics to optimize the design of the drone cluster stealth track planning based on task allocation. First, the comprehensive cost function of the drone cluster stealth track planning based on task allocation is designed; then on this basis, the drone cluster comprehensive cost function is minimized as the optimization goal, and the parameters such as the flight heading angle of the track are optimized to meet the track feasibility judgment, the drone cluster task allocation requirements and the drone cluster dynamics restrictions. Optimize the design, and construct a drone cluster stealth track planning optimization model based on task allocation; finally, the above optimization model is solved based on genetic algorithm and improved A* algorithm, the drone cluster task allocation scheme is optimized, and the cluster track is optimized, so as to achieve the purpose of improving the cluster track stealth performance and reducing the resource consumption of the drone cluster. By solving the optimization model, it is obtained that under the constraints of meeting the drone cluster task allocation requirements, track feasibility judgment and drone cluster dynamics restrictions, the drone cluster task allocation scheme x is mn and the heading angle of the drone is the optimal solution of the model.

Claims

1. An optimization design method for stealth trajectory planning of UAV cluster based on task allocation, characterized in that: The following steps are involved: Based on the range cost of the UAV trajectory, the threat level of the networked radar and the threat level of the anti-aircraft gun, a comprehensive cost function for the stealth trajectory planning of the UAV swarm based on task allocation is designed; Taking minimizing the comprehensive cost function of UAV cluster stealth trajectory planning based on task allocation as the optimization goal, and satisfying the trajectory feasibility judgment, UAV cluster task allocation plan requirements and UAV cluster dynamics constraints as constraints, an optimization model for UAV cluster stealth trajectory planning based on task allocation is constructed. An improved algorithm based on genetic algorithm and improved A* algorithm is used to solve the optimization model and obtain the optimal task allocation plan for the UAV cluster.

2. The method for optimizing the design of stealth trajectory planning of unmanned aerial vehicle clusters based on task allocation according to claim 1 is characterized in that: The expression of the comprehensive cost function of UAV cluster stealth trajectory planning based on task allocation is: in, is the comprehensive cost function of stealth trajectory planning of UAV swarm based on task allocation, M is the number of UAVs in the UAV swarm; represents the flight cost of the mth UAV track, which is evaluated by the UAV flight range, K m is the number of track segments contained in the track of the mth UAV, l m,k is the distance of the kth track segment of the mth UAV; is the threat level of the networked radar to the track of the mth UAV, which is evaluated by the probability and value of radar detection received by the track point; T m is the flight time of the mth drone; is the probability of the mth UAV being detected by the networked radar at time t; is the degree of anti-aircraft gun threat to the track of the mth UAV, which is evaluated by the probability and probability of anti-aircraft gun threat to the track point; is the probability that the mth UAV is threatened by anti-aircraft guns at time t; ω1, ω2, and ω3 are the weights of the range, the threat level of the networked radar, and the threat level of the anti-aircraft guns, respectively.

3. The method for optimizing the design of stealth trajectory planning of UAV cluster based on task allocation according to claim 1 is characterized in that: The expression of the optimization model of UAV cluster stealth trajectory planning based on task allocation is: in, is the comprehensive cost function of UAV cluster stealth trajectory planning based on task allocation, M is the number of UAVs in the UAV cluster, K m is the number of track segments contained in the track of the mth UAV, l m,k is the distance of the kth track segment of the mth UAV, T m is the flight time of the mth drone, is the probability of the mth UAV being detected by the networked radar at time t; is the probability that the mth UAV is threatened by anti-aircraft guns at time t; ω1, ω2, and ω3 are the weights of the range, the threat level of the networked radar, and the threat level of the anti-aircraft guns, respectively; L max is the maximum range of the drone, is the heading angle of the mth UAV at time t, is the heading angle of the mth UAV at time t+1, △ψ max is the maximum turning angle of each UAV; P c is the detection probability threshold of the networked radar, P a is the probability threshold of anti-aircraft gun detection, is the spatial position of the mth UAV at time t, obstacle is the spatial range of the fixed obstacle, σ is the RCS of the UAV according to the angle distribution, and σ q is the RCS of the drone at the qth angle, is the spatial position of the jth UAV at time t, d safe is the safety distance, x mn is the UAV cluster task allocation scheme, m is the UAV number, n is the task number, and N is the total number of tasks in the UAV cluster.

4. The method for optimizing the design of stealth trajectory planning of unmanned aerial vehicle clusters based on task allocation according to claim 1 is characterized in that: The improved algorithm based on genetic algorithm and improved A* algorithm is used to solve the optimization model, including: first, the improved A* algorithm is used to calculate the comprehensive track cost of each UAV to perform different tasks, and the comprehensive cost and track corresponding to each UAV to perform different tasks are stored in table S; then, based on table S, the genetic algorithm is used to solve the optimal task allocation plan for the UAV cluster.

5. The method for optimizing the design of stealth trajectory planning of unmanned aerial vehicle clusters based on task allocation according to claim 1 is characterized in that: The heuristic function expression used in the improved A* algorithm is: f(n0)=g(n0)+ωh(n0) Among them, f(n0) is the comprehensive cost function from the starting point to the end point through the track point n0; g(n0) is the actual cost function from the starting point to the track point n0; h(n0) is the estimated cost function from the track point n0 to the end point of the track, represented by the Manhattan distance; ω is the weighting factor.

6. The method for optimizing the design of stealth trajectory planning of unmanned aerial vehicle clusters based on task allocation according to claim 5 is characterized in that: The actual cost function g(n0) of the drone from the starting point to the track point n0 is expressed as: in, The actual flight distance of the mth UAV from the starting point to the track point n0, l m,k is the distance of the kth track segment of the mth UAV, is the number of track segments from the starting point to the track point n0 of the mth UAV, is the time taken by the mth UAV to travel from the starting point to the track point n0, It represents the probability of network radar detection from the starting point to the track point n0, is the probability that the mth UAV is detected by the networked radar at time t, represents the probability and threat probability of anti-aircraft guns on the track from the starting point to the track point n0. is the probability that the mth UAV is threatened by anti-aircraft guns at time t, ω1, ω2, and ω3 are the weights of the range, the threat level of the networked radar, and the threat level of the anti-aircraft guns, respectively.

7. The method for optimizing the design of stealth trajectory planning of unmanned aerial vehicle clusters based on task allocation according to claim 1 is characterized in that: The fitness function in the genetic algorithm is expressed as the inverse of the comprehensive cost of the cluster under different task allocation schemes.

8. The UAV cluster stealth trajectory planning optimization design system based on task allocation is characterized by: include: The comprehensive cost function design unit is used to design a comprehensive cost function for the stealth trajectory planning of UAV clusters based on task allocation, based on the range cost of the UAV trajectory, the threat level of the networked radar and the threat level of the anti-aircraft gun; The optimization model building unit is used to take minimizing the comprehensive cost function of the UAV cluster stealth trajectory planning based on task allocation as the optimization goal, and to meet the trajectory feasibility determination, the requirements of the UAV cluster task allocation plan and the UAV cluster dynamics constraints as the constraints, to build an optimization model for the UAV cluster stealth trajectory planning based on task allocation; The optimization model solving unit is used to solve the optimization model by adopting an improved algorithm based on genetic algorithm and improved A* algorithm to obtain the optimal task allocation plan for the drone cluster.

9. An electronic device, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the steps of the method for optimizing the design of stealth trajectory planning of a swarm of unmanned aerial vehicles based on task allocation as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which, when called, are used to execute the steps of the method for optimizing the design of stealth trajectory planning of a swarm of unmanned aerial vehicles based on task allocation as described in any one of claims 1 to 7.

Images