Multi-unmanned aerial vehicle cooperative target tracking path planning method based on spider wasp optimization algorithm

Abstract

The invention belongs to the technical field of unmanned aerial vehicle cluster control, and particularly relates to a multi-unmanned aerial vehicle cooperative target tracking path planning method based on a spider wasp optimization algorithm. Comprising the following steps: (1) establishing a mathematical constraint model of a flight path; (2) constructing an evaluation function according to the mathematical constraint model, wherein the evaluation function comprises a fitness function and a total voyage length; (3) based on a spider peak optimization algorithm, solving an evaluation function, and continuously iterating, calculating and comparing to obtain an optimal solution or an approximate optimal solution, namely an optimal path; and (4) distributing the optimal path to each unmanned aerial vehicle for execution. According to the method, efficient path planning of cooperative target tracking of multiple unmanned aerial vehicles in a complex three-dimensional environment is realized by simulating predation behaviors of spiders and social behaviors of swarms, the flight efficiency and safety of an unmanned aerial vehicle cluster are improved, and the adaptive capacity of the unmanned aerial vehicle cluster in the complex environment is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle cluster control, and in particular relates to a multi-unmanned aerial vehicle collaborative target tracking path planning method based on a spider wasp optimization algorithm. Background Art

[0002] Drones have the characteristics of strong adaptability, low cost, and strong scalability, and are now favored by all walks of life. Nowadays, with the continuous complexity of the external environment and the increase in mission types, it is difficult for a single drone to be competent for complex flight missions. Multi-drone collaboration has the advantages of distributed parallel perception, better fault tolerance and robustness, and improved mission execution efficiency. It has now been applied to various fields. However, multi-drone collaboration is a challenging research field, and there are many problems that need to be solved. The task allocation and path planning of drones in complex environments are of great significance for applications in various fields. Only through effective collaborative control can the drone system achieve maximum efficiency and maintain efficient and safe mission execution in the face of complex challenges.

[0003] Multi-UAV collaborative path planning refers to planning a set of effective paths for multiple UAVs when they work together. This path can ensure that UAVs can complete their tasks without collisions and work together in the best way. It can also help UAVs optimize routes, save energy, and improve efficiency, thereby better achieving task execution. However, with the increase in the number of UAVs, the increase in the complexity of task allocation, and the changes in the flight environment of UAVs, the current multi-UAV collaborative path planning still has problems such as slow convergence, easy to fall into local optimality, easy to cause collisions between machines, and insufficient path optimization. Therefore, it is necessary to focus on the research of multi-UAV collaborative path planning algorithms to improve the efficiency and safety of path planning. Summary of the invention

[0004] The purpose of the present invention is to provide a path planning method for multi-UAV cooperative target tracking based on spider-wasp optimization algorithm. The method simulates the predation behavior of spiders and the social behavior of bee colonies to achieve efficient path planning for cooperative target tracking of multiple UAVs in complex three-dimensional environments, improve the flight efficiency and safety of UAV clusters, and enhance the adaptability of UAV clusters in complex environments.

[0005] The technical solution to achieve the purpose of the present invention is: a multi-UAV collaborative target tracking path planning method based on the spider wasp optimization algorithm, characterized in that it includes the following steps:

[0006] Step (1): Establish a mathematical constraint model of the flight path;

[0007] Step (2): construct an evaluation function based on the mathematical constraint model, including the fitness function and the total flight length;

[0008] Step (3): Based on the spider peak optimization algorithm, solve the evaluation function, iterate, calculate, and compare continuously to obtain the optimal solution or approximate optimal solution, that is, the optimal path;

[0009] Step (4): Distribute the optimal path to each drone for execution.

[0010] 2. The method according to claim 1, characterized in that step (1) comprises the following steps:

[0011] Step (11): Assume that the path length of the drone from track point i to track point i+1 is x i , the maximum range that the drone can fly is x max , then the length constraint for the drone to complete the assigned task is:

[0012]

[0013] Step (12): When dealing with the threat model of obstacles, simplify the obstacles into circles with the center coordinates of the circle (x, y), the threat radius R, and the current position of the drone (x k ,y k ), the next location to be flown to is (x k+1 ,y k+1 ), draw a vertical line from the center of the obstacle circle to intersect at point (x0, y0), the distance between these two points is the shortest safe distance for the drone to avoid obstacles; introduce the expansion radius r on the basis of the threat radius R, so that the value of R+r is greater than or equal to the shortest safe distance. For a single drone, the obstacle threat coefficient formula in the flight path is:

[0014]

[0015] Among them, α represents the threat weight, which takes values ​​in (0,1) to indicate the potential threat level, n is the number of path points during the flight of the drone, and r is the expansion radius of the obstacle; if h i If the value is positive, the drone can safely avoid threatening obstacles; otherwise, the drone may collide with threatening obstacles when flying to the next path point and needs to select a new path point again;

[0016] Step (13): When the drone flies from the current path point to the next path point, if the two path points are not in the same plane, the pitch angle of the drone must be changed; (x k ,y k ,z k ) is the position coordinate of the drone at time k, (x k+1 ,y k+1 ,z k+1) is the position coordinate of the UAV at time k+1, θ is the pitch angle of the UAV; suppose the UAV is at (x k ,y k ,z k ) at the maximum elevation or depression angle to reach the next path point safely and stably. The coordinates are (x k+1 ,y k+1 ,z k+1 ), then the pitch angle threshold range of the safe stable point is calculated by the following formula:

[0017]

[0018] Step (14): Assume the position of the drone at point A is (x k-1 ,y k-1 ,z k-1 ), the position of point B is (x k ,y k ,z k ), the position of point C is (x k+1 ,y k+1 ,z k+1 ), the position of point D is (x1, y1, z1), and the position of point E is (x2, y2, z2), where point D and point E are the two endpoints of the obstacle diameter. is the maximum turning angle of the drone, is the maximum turning angle of the drone, and the calculation formula for the turning angle range is:

[0019]

[0020] Step (15): Set time constraints for drones that perform different tasks at the same target point. The constraint expressions are as follows:

[0021] t a +t min ≤t b

[0022] Where t a The time it takes for the first UAV to reach the target point and complete the assigned task, t min is the minimum time interval between two UAVs reaching the same target point, t b is the time when the second UAV reaches the target point.

[0023] 3. The method according to claim 2, characterized in that step (2) specifically comprises the following steps:

[0024] Step (21): Establish a fitness function: The fitness function is composed of the maximum flight time of the drone, the average flight range of the drone, and the value of the approach relationship between the drone and the obstacle during the flight, as follows:

[0025]

[0026] a+b=1

[0027] Where: h N is the number of drones, h U is the number of obstacles, x j is the flight distance of UAV j, p j (k+1) is the position coordinate of UAV j at the k+1th moment, obs(u) is the position of the uth obstacle, λ is the threat coefficient, a and b represent the weight coefficients of flight time and fuel consumption respectively, and v0 is the flight speed of the UAV;

[0028] Step (22): Establish the total flight length: The total flight length is composed of the flight length of all drones and the value of the approach relationship between the drone and the obstacle during the flight, as follows:

[0029]

[0030] 4. The method according to claim 3, characterized in that step (3) specifically comprises the following steps:

[0031] Step (31): Initialization: randomly generate a set of initial positions and velocities of the drones, as well as the position information of the target; at the same time, set the control parameters of the algorithm;

[0032] Each female spider wasp represents a solution for the current generation, encoded as a D-dimensional vector using the following expression:

[0033]

[0034] A set of N vectors is randomly generated within the predefined upper and lower initial parameter bounds.

[0035]

[0036] Among them, SW Pop is the initial population of spider wasps;

[0037] The following formula is used to randomly generate any solution in the search space:

[0038]

[0039] Where t is the generated index, i is the overall index (i = 1, 2, ..., N), is a vector of D-dimensional random initialization numbers between 0 and 1;

[0040] Step (32): Evaluate fitness: Calculate the fitness value, i.e., the objective function value of path planning, based on the position and speed of each UAV;

[0041] Step (33): Update position and speed: Update the position and speed of each drone according to the behavior strategy of the spider wasp;

[0042] Step (34): Evaluate the updated fitness: Recalculate the fitness value of each drone based on the updated position and speed:

[0043] N=N min +(NN min )×k

[0044] Step (35): Update the optimal solution: Update the global optimal solution and the individual optimal solution according to the fitness value; the global optimal solution represents the position of the drone that is closest to the target and has the best path among all the current drones.

[0045] Step (3-6): Determine the termination condition: Check whether the preset number of iterations is reached or the stopping criterion is met; if the termination condition is met, output the optimal solution, that is, the optimal path; otherwise, continue iterating;

[0046] Step (37): Execute the path.

[0047] 5. The method according to claim 4, characterized in that step (33) specifically comprises:

[0048] The drone updates its position based on its current position and speed through local search and global search strategies to simulate the hunting and nesting behavior of spider wasps; it further optimizes its position through random search and elite search strategies to simulate the mating behavior of spider wasps;

[0049] Search phase:

[0050] Randomly explore the search space with a constant step size:

[0051]

[0052] μ1=|rn|*r1

[0053] Search for key areas:

[0054]

[0055] μ2=B*cos(2πl)

[0056]

[0057] The two search methods complement each other to explore the search space and locate the most promising areas:

[0058]

[0059] r3, r4 are random numbers in [0,1];

[0060] Follow and flee phase:

[0061] Keep following the optimal solution,

[0062]

[0063] is a random vector in [0,1], r6 is a random number in [0,1];

[0064] When gradually moving away from the optimal solution, as the distance increases, following is transformed into continuing the search:

[0065]

[0066] is a vector generated between k and -k according to a normal distribution:

[0067]

[0068] The trade-off between the two trends is realized randomly,

[0069]

[0070] At the beginning of the optimization algorithm, a global search is performed to explore and exploit key areas during the iteration process to avoid getting stuck in local minima; the search phase is adjusted according to the following formula:

[0071]

[0072] p is a random number in [0,1];

[0073] Nesting behavior:

[0074]

[0075] Mating behavior:

[0076]

[0077] Compared with the prior art, the present invention has the following significant advantages:

[0078] (1) Significantly improve flight efficiency: The path planning method for multi-UAV collaborative target tracking based on the spider-wasp optimization algorithm significantly reduces unnecessary flight distance and time by intelligently allocating and optimizing the flight paths of each UAV. The algorithm can efficiently find the global or near-global optimal path in a complex three-dimensional environment, avoiding detours or stagnation of UAVs during flight, thereby greatly improving the overall flight efficiency of the UAV cluster.

[0079] (2) Enhanced mission execution capability: In military reconnaissance, target tracking, material delivery and other tasks, drone clusters need to quickly and accurately reach designated locations and perform tasks. The path planning system of the present invention can ensure that drone clusters can still collaborate efficiently in complex environments and quickly respond to mission requirements, thereby enhancing the mission execution capability of drone clusters.

[0080] (3) Improving flight safety: In a three-dimensional environment, drones face challenges from multiple obstacles such as undulating terrain, buildings, and trees. The algorithm of the present invention fully considers the need for obstacle avoidance in path planning. Through intelligent search and evaluation, it ensures that the generated path can safely bypass all obstacles, avoiding the risk of collision between drones and obstacles, thereby improving flight safety.

[0081] (4) Enhanced system robustness: The spider-wasp optimization algorithm combines the advantages of global search and local search, and can dynamically adjust the search strategy during the search process to cope with changes in different environments and task requirements. This flexible search mechanism enables the drone cluster to quickly adjust the path planning plan when faced with emergencies or environmental changes, thereby maintaining the stability and robustness of the system.

[0082] (5) Reduce energy consumption and costs: By optimizing the flight path and reducing unnecessary flight distance and time, the present invention can significantly reduce the energy consumption and operating costs of drone clusters; this has very important economic significance for application scenarios that require long-term and large-scale deployment of drone clusters. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a flow chart of the path planning method of the present invention.

[0084] Figure 2 This is a flow chart of the spider wasp optimization algorithm of the present invention.

[0085] Figure 3 Schematic diagram of multi-UAV path planning scheme.

[0086] Figure 4 This is a bird’s-eye view of the multi-UAV path planning solution.

[0087] Figure 5is the total cost of all UAV path planning.

[0088] Figure 6 is the path planning cost of each UAV. DETAILED DESCRIPTION

[0089] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0090] A path planning method for multi-UAV cooperative target tracking based on spider-wasp optimization algorithm, comprising the following steps:

[0091] Step (1): Establish mathematical constraint models such as navigation length, track, pitch angle, horizontal turning angle and time.

[0092] Step (1-1): Assume that the path length from track point i to track point i+1 is x i , the maximum range that the drone can fly is x max , then the length constraint for the drone to complete the assigned task is:

[0093]

[0094] Step (1-2): The trajectory constraint model is the obstacle threat model. When dealing with the obstacle threat model, it is necessary to maintain the stability of the drone's flight altitude and the overhead viewing angle. The obstacle is simplified into a circle with the center coordinate of the circle (x, y) and the threat radius R. The drone's current position (x k ,y k ), the next location to be flown to is (x k+1 ,y k+1 ), draw a vertical line from the center of the obstacle circle to intersect at point (x0, y0), and the distance between these two points is the shortest safe distance for the drone to avoid obstacles. To ensure that the drone has enough safe distance, an expansion radius r is introduced on the basis of the obstacle radius, so that the value of R+r is greater than or equal to the shortest safe distance, thereby further improving the safety of drone flight. For a single drone, the obstacle threat coefficient formula in its flight path is:

[0095]

[0096] Among them, α represents the threat weight, which takes values ​​in (0,1) to indicate the potential threat level, n is the number of path points during the flight of the drone, and r is the expansion radius of the obstacle. i If the value is positive, the drone can safely avoid threatening obstacles; otherwise, the drone may collide with threatening obstacles when flying to the next path point and needs to select a new path point again.

[0097] Step (1-3): When the drone flies from the current path point to the next path point, if the two path points are not in the same plane, the pitch angle of the drone must be changed. k ,y k ,z k ) is the position coordinate of the drone at time k, (x k+1 ,y k+1 ,z k+1 ) is the position coordinate of the UAV at time k+1, and θ is the pitch angle of the UAV. k ,y k ,z k ) at the maximum elevation or depression angle to safely and stably reach the next path point with a position coordinate of (x k+1 ,y k+1 ,z k+1 ), the pitch angle threshold range of the safe stable point can be calculated by the following formula:

[0098]

[0099] Step (1-4): In order to avoid obstacles, the drone needs to turn during flight. If the turning angle is too large, the drone will become unbalanced; if the turning angle is too small, it will not be able to avoid obstacles. Suppose the drone is at point A (x k-1 ,y k-1 ,z k-1 ), the position of point B is (x k ,y k ,z k ), the position of point C is (x k+1 ,y k+1 ,z k+1 ), the position of point D is (x1, y1, z1), the position of point E is (x2, y2, z2), (point D and point E are the two endpoints of the obstacle diameter), is the maximum turning angle of the drone, is the maximum turning angle of the drone, and the calculation formula for the turning angle range is:

[0100]

[0101] Steps (1-5): When multiple drones work together to complete the assigned target tasks, there may be certain constraints. For example, drones need to complete three tasks: reconnaissance, strike, and review. The drone that performs the strike task should go to the same target location to perform the task after the drone that completes the reconnaissance task. At the same target location, collisions between drones must be prevented. Therefore, time constraints must be set for drones that perform different tasks at the same target point. The constraint expressions are as follows:

[0102] ta +t min ≤t b

[0103] Where t a The time it takes for the first UAV to reach the target point and complete the assigned task, t min is the minimum time interval between two UAVs reaching the same target point, t b is the time when the second UAV reaches the target point.

[0104] Step (2): Construct an evaluation function based on the mathematical constraint model, including the fitness function and the total flight length.

[0105] Step (2-1): Establish a fitness function. The fitness function consists of three parts: the maximum flight time of the drone, the average flight distance of the drone, and the cost of the approach relationship between the drone and obstacles during flight.

[0106]

[0107] a+b=1

[0108] Where: h N is the number of drones, h U is the number of obstacles, x j is the flight distance of UAV j, p j (k+1) is the position coordinate of UAV j at the k+1th moment, obs(u) is the position of the uth obstacle, λ is the threat coefficient, a and b represent the weight coefficients of flight time and fuel consumption respectively, and v0 is the flight speed of the UAV.

[0109] Step (2-2): Establish the total flight length. The total flight length consists of two parts: the flight length of all drones and the cost value of the approach relationship between the drone and the obstacle during flight.

[0110]

[0111] Step (3): Based on the spider-wasp optimization algorithm, solve the evaluation function, continuously iterate, calculate, and compare to obtain the optimal solution or approximate optimal solution, which is the three-dimensional path planning solution for the drone cluster.

[0112] Step (3-1): Initialization: Randomly generate a group of initial positions and velocities of drones, as well as the position information of the target. At the same time, set the control parameters of the algorithm, such as population size, maximum number of iterations, crossover rate, etc.

[0113] Each spider wasp (female) represents a solution of the current generation and can be encoded as a D-dimensional vector using the following expression:

[0114]

[0115] A set of N vectors is randomly generated within the predefined upper and lower initial parameter bounds.

[0116]

[0117] Among them, SW Pop is the initial population of spider wasps.

[0118] The following formula is used to randomly generate any solution in the search space:

[0119]

[0120] Where t is the generated index, i is the overall index (i = 1, 2, ..., N), is a vector of D-dimensional random initialization numbers between 0 and 1.

[0121] Step (3-2): Evaluate fitness: According to the position and speed of each drone, calculate its fitness value, that is, the objective function value of path planning.

[0122] Step (3-3): Update position and speed: Update the position and speed of each drone according to the behavior strategy of the spider wasp. Specifically, the drone updates its position based on its current position and speed through local search and global search strategies to simulate the hunting and nesting behavior of the spider wasp. At the same time, the position is further optimized through random search and elite search strategies to simulate the mating behavior of the spider wasp.

[0123] Search phase:

[0124] Randomly explore the search space with a constant step size:

[0125]

[0126] μ1=|rn|*r1

[0127] Search for key areas:

[0128]

[0129] μ2=B*cos(2πl)

[0130]

[0131] The two search methods complement each other to explore the search space and locate the most promising areas.

[0132]

[0133] r3, r4 are random numbers in the range [0,1].

[0134] Follow and flee phase:

[0135] Keep following the optimal solution,

[0136]

[0137] is a random vector in [0,1], and r6 is a random number in [0,1].

[0138] When gradually moving away from the optimal solution, as the distance increases, following is transformed into continued searching.

[0139]

[0140] is a vector generated between k and -k according to a normal distribution.

[0141]

[0142] The trade-off between the two trends is realized randomly,

[0143]

[0144] At the beginning of the optimization algorithm, a global search is performed to explore and exploit key areas during the iteration process to avoid falling into local minima. Adjust the search phase according to the following formula:

[0145]

[0146] p is a random number in [0,1].

[0147] Nesting behavior:

[0148]

[0149] Mating behavior:

[0150]

[0151] Step (3-4): Evaluate the updated fitness: Recalculate the fitness value of each drone based on the updated position and speed.

[0152] N=N min +(NN min )×k

[0153] Step (3-5): Update the optimal solution: Update the global optimal solution and the individual optimal solution according to the fitness value. The global optimal solution represents the position of the drone that is closest to the target and has the best path among all the current drones.

[0154] Step (3-6): Determine the termination condition: Check whether the preset number of iterations is reached or the stopping criteria are met. If the termination condition is met, the optimal solution, i.e. the optimal path, is output; otherwise, return to step 3 to continue iteration.

[0155] Step (3-7): Execution path: Based on the obtained optimal path, control multiple drones to collaboratively track the target.

[0156] Example 1

[0157] A more detailed description is given below in conjunction with embodiments.

[0158] The path planning solution given by the algorithm proposed in this paper is verified to be reasonable and effective in a simulated environment. Ten obstacle threat zones are established in the map, and no flight is allowed at any altitude in the threat zone. The size of the simulation map is set to 1000m×1000m×200m. The starting point coordinates are (150, 150, 50), and the end point coordinates are (900, 720, 50).

[0159] The specific settings of obstacle threat areas are shown in Table 1 below.

[0160] Table 1 Obstacle threat zone settings

[0161]

[0162] The algorithm parameter configuration is shown in Table 2 below.

[0163] Table 2 Algorithm parameter configuration

[0164]

[0165]

[0166] In this simulation experiment, the number of drones is set to 5, the population size is set to 100, and the maximum number of iterations is set to 500.

[0167] According to the steps of the present invention, the path planning is performed, and the optimal path finally obtained is as follows: Figure 3 and Figure 4 shown.

[0168] In summary, the path planning method for multi-UAV cooperative target tracking based on the spider wasp optimization algorithm achieves fast and accurate planning of UAV paths by simulating the behavior of spider wasps. This method has the advantages of strong adaptability, high robustness, and high solution accuracy, and provides a new solution for UAV cooperative target tracking.

Claims

1. A multi-UAV cooperative target tracking path planning method based on the spider-wasp optimization algorithm, characterized in that: The steps include: Step (1): Establish a mathematical constraint model of the flight path; Step (2): construct an evaluation function based on the mathematical constraint model, including the fitness function and the total flight length; Step (3): Based on the spider peak optimization algorithm, solve the evaluation function, iterate, calculate, and compare continuously to obtain the optimal solution or approximate optimal solution, that is, the optimal path; Step (4): Distribute the optimal path to each drone for execution.

2. The method according to claim 1, characterized in that Step (1) comprises the following steps: Step (11): Assume that the path length of the drone from track point i to track point i+1 is x i , the maximum range that the drone can fly is x max , then the length constraint for the drone to complete the assigned task is: Step (12): When dealing with the threat model of obstacles, simplify the obstacles into circles with the center coordinates of the circle (x, y), the threat radius R, and the current position of the drone (x k ,y k ), the next location to be flown to is (x k+1 ,y k+1 ), draw a vertical line from the center of the obstacle circle to intersect at point (x0, y0), the distance between these two points is the shortest safe distance for the drone to avoid obstacles; introduce the expansion radius r on the basis of the threat radius R, so that the value of R+r is greater than or equal to the shortest safe distance. For a single drone, the obstacle threat coefficient formula in the flight path is: Among them, α represents the threat weight, which takes values ​​in (0,1) to indicate the potential threat level, n is the number of path points during the flight of the drone, and r is the expansion radius of the obstacle; if h i If the value is positive, the drone can safely avoid threatening obstacles; otherwise, the drone may collide with threatening obstacles when flying to the next path point and needs to select a new path point again; Step (13): When the drone flies from the current path point to the next path point, if the two path points are not in the same plane, the pitch angle of the drone must be changed; (x k ,y k ,z k ) is the position coordinate of the UAV at time k, (x k+1 ,y k+1 ,z k+1 ) is the position coordinate of the UAV at time k+1, θ is the pitch angle of the UAV; suppose the UAV is at (x k ,y k ,z k ) at the maximum elevation or depression angle to reach the next path point safely and stably. The coordinates are (x k+1 ,y k+1 ,z k+1 ), then the pitch angle threshold range of the safe stable point is calculated by the following formula: Step (14): Assume the position of the drone at point A is (x k-1 ,y k-1 ,z k-1 ), the position of point B is (x k ,y k ,z k ), the position of point C is (x k+1 ,y k+1 ,z k+1 ), the position of point D is (x1, y1, z1), and the position of point E is (x2, y2, z2), where point D and point E are the two endpoints of the obstacle diameter. is the maximum turning angle of the drone, is the maximum turning angle of the drone, and the calculation formula for the turning angle range is: Step (15): Set time constraints for drones performing different tasks at the same target point. The constraint expressions are as follows: t a +t min ≤t b Where t a The time it takes for the first UAV to reach the target point and complete the assigned task, t min is the minimum time interval between two UAVs reaching the same target point, t b is the time when the second UAV reaches the target point.

3. The method according to claim 2, characterized in that Step (2) specifically includes the following steps: Step (21): Establish a fitness function: The fitness function is composed of the maximum flight time of the drone, the average flight range of the drone, and the value of the approach relationship between the drone and the obstacle during the flight, as follows: a+b=1 Where: h N is the number of drones, h U is the number of obstacles, x j is the flight distance of UAV j, p j (k+1) is the position coordinate of UAV j at the k+1th moment, obs(u) is the position of the uth obstacle, λ is the threat coefficient, a and b represent the weight coefficients of flight time and fuel consumption respectively, and v0 is the flight speed of the UAV; Step (22): Establish the total flight length: The total flight length is composed of the flight length of all drones and the value of the approach relationship between the drone and the obstacle during the flight, as follows:

4. The method according to claim 3, characterized in that: Step (3) specifically includes the following steps: Step (31): Initialization: randomly generate a set of initial positions and velocities of the drones, as well as the position information of the target; at the same time, set the control parameters of the algorithm; Each female spider wasp represents a solution for the current generation, encoded as a D-dimensional vector using the following expression: A set of N vectors is randomly generated within the predefined upper and lower initial parameter bounds. Among them, SW Pop is the initial population of spider wasps; The following formula is used to randomly generate any solution in the search space: Where t is the generated index, i is the overall index (i = 1, 2, ..., N), is a vector of D-dimensional random initialization numbers between 0 and 1; Step (32): Evaluate fitness: Calculate the fitness value, i.e., the objective function value of path planning, based on the position and speed of each UAV; Step (33): Update position and speed: Update the position and speed of each drone according to the behavior strategy of the spider wasp; Step (34): Evaluate the updated fitness: Recalculate the fitness value of each drone based on the updated position and speed: N=N min +(N-N min )×k Step (35): Update the optimal solution: Update the global optimal solution and the individual optimal solution according to the fitness value; the global optimal solution represents the position of the drone that is closest to the target and has the best path among all the current drones. Step (3-6): Determine the termination condition: Check whether the preset number of iterations is reached or the stopping criterion is met; if the termination condition is met, output the optimal solution, that is, the optimal path; otherwise, continue iterating; Step (37): Execute the path.

5. The method according to claim 4, characterized in that Step (33) is specifically: The drone updates its position based on its current position and speed through local search and global search strategies to simulate the hunting and nesting behavior of spider wasps; it further optimizes its position through random search and elite search strategies to simulate the mating behavior of spider wasps; Search phase: Randomly explore the search space with a constant step size: μ1=|rn|*r1 Search for key areas: μ2=B*cos(2πl) The two search methods complement each other to explore the search space and locate the most promising areas: r3, r4 are random numbers in [0,1]; Follow and flee phase: Keep following the optimal solution, r5 is a random vector in [0,1], r6 is a random number in [0,1]; When gradually moving away from the optimal solution, as the distance increases, following is transformed into continuing the search: is a vector generated between k and -k according to a normal distribution: The trade-off between the two trends is realized randomly, At the beginning of the optimization algorithm, a global search is performed to explore and exploit key areas during the iteration process to avoid getting stuck in local minima; the search phase is adjusted according to the following formula: p is a random number in [0,1]; Nesting behavior: Mating behavior:

Images