Cluster unmanned aerial vehicle defense penetration path planning method and system, memory and electronic equipment

Abstract

The invention provides a cluster unmanned aerial vehicle defense penetration path planning method and system, a memory and electronic equipment, and belongs to the technical field of unmanned aerial vehicle control, and the method comprises the steps: S1, obtaining a path point solution set of a cluster unmanned aerial vehicle; s2, introducing a dynamic weight coefficient into a particle swarm algorithm, and constructing an improved particle swarm algorithm; s3, substituting the path point solution set into an improved particle swarm algorithm, and determining the position of each path point; s4, constructing a fitness function related to the positions of the path points, substituting the position of each path point into the fitness function, and calculating an optimal solution under the current iteration frequency; and S5, repeating the steps S2-S4 until the maximum number of iterations is reached, determining a globally optimal solution, and taking a path formed by path points corresponding to the globally optimal solution as a penetration path. According to the method, the dynamic weight coefficient is introduced into the particle swarm algorithm, the dynamic weight coefficient is continuously updated in the iteration process, the current space is searched, the global search capability of the cluster unmanned aerial vehicle is improved, and falling into a local optimal solution is avoided.

Description

Technical Field

[0001] The present invention belongs to the field of UAV control technology, and relates to a path planning technology for cluster UAVs, and specifically to a cluster UAV penetration path planning method, system, memory and electronic equipment. Background Art

[0002] Low-altitude penetration is a tactical means. By maintaining a relatively low flight altitude, the UAV can make full use of terrain shielding, reduce the target cross-sectional area, and minimize the possibility of exposure. Low-altitude penetration can ensure the successful execution of the mission, improve survivability, and avoid enemy air defense threats.

[0003] Swarm drones have the advantages of strong fault tolerance, good adaptive coordination, and high cluster intelligence. If there is an emergency in the actual flight of swarm drones, the track must be replanned to avoid threats, that is, penetration path planning. In order to meet the timeliness of drones working together, the algorithm used for replanning must be real-time and efficient. Therefore, during the flight, the drone needs to continuously correct the reference track according to the threat information obtained until it reaches the target node. Refer to the patent application document with application number CN202211422410.6, which discloses a collaborative path planning method in a multi-threat environment, the method comprising: planning all feasible threat avoidance paths according to the starting position and the end position of each drone in the drone cluster; determining the basic path in the feasible threat avoidance path of each drone according to the expected path length, and optimizing and adjusting the determined basic path so that the length of the optimized basic path and the optimized length of the expected path length are minimized, and the first n optimized basic paths arranged from low to high in the optimized length error are used as the optimized path set; performing collision detection on the optimized path set of each drone, and using the detected collision-free optimized path as the collaborative path. It can be seen from this that the prior art has conducted certain research on the path planning of cluster drones in multiple threat environments, but the existing cluster drone path planning does not consider the entire search space, resulting in poor global search capabilities of drones and easy to fall into local optimality. Summary of the invention

[0004] In response to the technical problem described in the above background technology, the existing swarm UAV path planning does not take the entire search space into consideration, resulting in poor global search capabilities of the UAVs and easy trapping in local optimality. To address this technical problem, the present invention proposes a swarm UAV penetration path planning method, system, memory and electronic device.

[0005] The present invention introduces a dynamic weight coefficient into the particle swarm algorithm, continuously updates the dynamic weight coefficient during the iteration process, and gradually adjusts the dynamic weight coefficient, thereby affecting the movement behavior of the path point. During each iteration, it is necessary to fully search the current space, thereby improving the global search capability of the UAV and preventing the UAV's penetration path from falling into a local optimal solution.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The cluster UAV penetration path planning method of the present invention comprises the following steps:

[0008] S1: Obtain the path point solution set of the swarm drones;

[0009] S2: Introduce dynamic weight coefficient into the particle swarm algorithm to construct an improved particle swarm algorithm;

[0010] S3: Substitute the path point solution set into the improved particle swarm algorithm to determine the position of each path point;

[0011] S4: Construct a fitness function F related to the location of the path point itness , substitute the position of each path point into the fitness function F itness Calculate the optimal solution under the current number of iterations

[0012] S5: Repeat steps S2-S4 until the maximum number of iterations is reached and the global optimal solution X is determined. g , the global optimal solution X g The path formed by the corresponding path points serves as the penetration path of the swarm UAVs.

[0013] It is further defined that in step S2, the improved particle swarm algorithm is:

[0014]

[0015] In the formula, is the velocity of the i-th path point at the t-th iteration; ω t is the dynamic weight coefficient at the tth iteration; is the speed of the i-th path point at the (t-1)th iteration; is the cognitive coefficient at the (t-1)th iteration; a t-1 is the chaos coefficient at the (t-1)th iteration; X ip is the optimal solution of the i-th path point in history; is the optimal solution of the i-th path point at the (t-1)th iteration; is the social coefficient at the (t-1)th iteration; is the optimal solution at the first iteration; is the optimal solution of the i-th path point at the t-th iteration; t is the current iteration number and is an integer ≥1.

[0016] Further defined, the expression of the dynamic weight coefficient is:

[0017] ω t =ω max -0.25*log10 t

[0018] In the formula, ω t is the dynamic weight coefficient at the tth iteration; ω max is the maximum dynamic weight coefficient, which is 1; t is the current iteration number.

[0019] It is further defined that in step S3, in the first iteration, the pseudo global optimal solution is calculated from the path point solution set using the reverse roulette algorithm The pseudo global optimal solution as the optimal solution for the first iteration.

[0020] It is further defined that step S4 comprises: constructing a fitness function F related to the position of the path point itness , and use the Meltrov criterion to determine the optimal solution under the current number of iterations

[0021] It is further defined that the fitness function F itness for:

[0022] F itness =F 1 +F 2 +F 3 +F 4

[0023] In the formula, F 1 is the terrain threat cost function; F 2 is the energy consumption cost function; F 3 is the radar threat cost function; F 4 is the flight altitude cost function; where F 1 and F 2 They are all functions related to the position of the path points.

[0024] It is further defined that the terrain threat cost function F 1 for:

[0025] F 1 =R des *K 1

[0026] In the formula, R desis the crash rate of drones in the swarm, which is a function related to the position of the path point; K 1 is a constant, with a value of 0.3;

[0027] The energy consumption cost function F 2 for:

[0028] F 2 =R e *K 2

[0029] In the formula, R e is the energy consumption rate, which is a function related to the position of the path point; K 2 is a constant, with a value of 0.2;

[0030] The radar threat cost function F 3 for:

[0031] F 3 =P det *K 3

[0032] Where P det is the probability of all radars being exposed; K 3 is a constant, and its value is 0.3.

[0033] The cluster UAV penetration path planning system of the present invention is applied to the above-mentioned cluster UAV penetration path planning method, comprising:

[0034] Acquisition module: used to obtain the path point solution set of the cluster drones;

[0035] Algorithm building module: used to introduce dynamic weight coefficients into the particle swarm algorithm and build an improved particle swarm algorithm;

[0036] Waypoint location determination module: used to substitute the waypoint solution set into the improved particle swarm algorithm to determine the location of each waypoint;

[0037] Calculation module: used to construct the fitness function F related to the location of the path point itness , substitute the position of each path point into the fitness function F itness Calculate the optimal solution under the current number of iterations

[0038] And iterative module: used to repeatedly iterate the algorithm construction module, the path point location determination module and the calculation module until the maximum number of iterations is reached to determine the global optimal solution X g , the global optimal solution X g The path formed by the corresponding path points serves as the penetration path of the swarm UAVs.

[0039] The present invention provides a memory storing a program file, wherein the program file is executed to implement program instructions formed by the above-mentioned cluster UAV penetration path planning method.

[0040] An electronic device of the present invention comprises a processor and a memory coupled to each other, wherein the memory is used to store program instructions formed by the above-mentioned cluster UAV penetration path planning method;

[0041] The processor is used to execute program instructions stored in the memory.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. The cluster UAV penetration path planning method of the present invention introduces a dynamic weight coefficient into the particle swarm algorithm, and continuously updates the dynamic weight coefficient during the iteration process, gradually adjusts the dynamic weight coefficient, thereby affecting the movement behavior of the path point. Each time iterates, it is necessary to fully search the current space, thereby improving the global search capability of the cluster UAV and preventing the cluster UAV's penetration path from falling into a local optimal solution.

[0044] 2. In the present invention, the Meltrov criterion is used in each generation of particle update to accept new states and suboptimal solutions with a certain probability, further preventing the penetration path of the swarm UAV from falling into the local optimal solution.

[0045] 3. In the present invention, in the first iteration, the reverse roulette algorithm is used to calculate the pseudo global optimal solution from the path point solution set. The pseudo global optimal solution As the optimal solution in the first iteration, it further prevents the penetration path of the swarm UAVs from falling into the local optimal solution.

[0046] 4. In the present invention, a chaotic distribution random factor is also added to the particle swarm algorithm. The chaotic distribution random factor can make the cluster UAVs have higher randomness and unpredictability in complex environments, so as to improve the guiding ability of the cluster UAVs in the global search process.

[0047] 5. In the present invention, the fitness function is formed by the terrain threat cost function, the energy consumption cost function, the radar threat cost function and the flight altitude cost function. The present invention fully considers the influence of various interference factors on the penetration path of cluster UAVs, and greatly reduces the risk of cluster UAVs in the penetration process. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 Schematic diagram of the penetration path planning method of swarm UAVs of the present invention;

[0049] Figure 2 A schematic diagram of a swarm UAV penetration path planning system of the present invention;

[0050] Figure 3 It is a hemispherical distribution ultraviolet MIMO structure model;

[0051] Figure 4 A schematic diagram of maintaining a UV-guided link between one-to-many UAVs;

[0052] Figure 5 This is the formation topology diagram of the swarm drones. DETAILED DESCRIPTION

[0053] The technical solution of the present invention is further explained below in conjunction with the accompanying drawings and embodiments, but the present invention is not limited to the implementation modes described below.

[0054] See also Figure 3 , which is a hemispherical distribution ultraviolet MIMO structure model, where MIMO is the abbreviation of MultiInputMultiOutput. Compared with the traditional planar distribution-based ultraviolet MIMO structure model, it is easier to achieve three-dimensional guidance coverage of the UAV flight space. The hemispherical distribution ultraviolet MIMO structure model of the present invention can achieve ultraviolet signal guidance coverage in three-dimensional space while minimizing signal interference of ultraviolet light guidance links between UAVs through ultraviolet LED time-division coding communication.

[0055] See also Figure 4 , is a schematic diagram of maintaining the ultraviolet light leading link between one-to-many UAVs of the present invention. The ultraviolet guiding light field of the ultraviolet MIMO structure model with hemispherical distribution is divided into three areas, namely, the warning area located in the central area, the identification area located in the edge area, and the guiding area located between the warning area and the identification area. The identification area is responsible for sensing the UAV outside the formation that acts as an obstacle. The UAV located in the warning area is the leading UAV, and the UAV located in the guiding area is the following UAV. The warning area is used to maintain the minimum safety distance between the leading UAV and the following UAV. When the leading UAV in the warning area and the following UAV in the guiding area try to establish an ultraviolet light leading link, the leading UAV sends a request and positioning information (its own node number and longitude and latitude information) to the central processor. The central processor generates a guiding instruction for the following UAV based on the UAV positioning information of the leading UAV and the UAV positioning information of the following UAV, under the condition of meeting the minimum safety distance, so that the leading UAV and the following UAV establish an ultraviolet light leading link. During the flight, the leading UAV sends motion status information to the following UAV to adjust its flight behavior in real time, ensuring that the ultraviolet light leading link between the one-to-many UAVs is maintained.

[0056] The present invention maintains the leading link and positioning of the cluster UAV (composed of the leading UAV and the following UAV) through the hemispherical distribution ultraviolet MIMO structural model, and introduces a dynamic weight coefficient in the particle swarm algorithm, and continuously updates the dynamic weight coefficient during the iteration process, thereby affecting the movement behavior of each path point. In each iteration, the hemispherical distribution ultraviolet MIMO structural model needs to fully search the current space, thereby improving the global search capability of the cluster UAV and preventing the cluster UAV's penetration path from falling into a local optimal solution.

[0057] Example 1

[0058] See also Figure 1 The cluster UAV penetration path planning method of this embodiment includes the following steps:

[0059] S1: Obtain the path point solution set of the swarm drones; the path point solution set is a set of all path points between the starting point of the swarm drones and the target point of the swarm drones.

[0060] S2: Introduce dynamic weight coefficient into the particle swarm algorithm and construct an improved particle swarm algorithm; the improved particle swarm algorithm is:

[0061]

[0062] In the formula, is the velocity of the i-th path point at the t-th iteration; ω t is the dynamic weight coefficient at the tth iteration; is the speed of the i-th path point at the (t-1)th iteration; is the cognitive coefficient at the (t-1)th iteration; a t-1 is the chaos coefficient at the (t-1)th iteration; X ip is the optimal solution of the i-th path point in history; is the optimal solution of the i-th path point at the (t-1)th iteration; is the social coefficient at the (t-1)th iteration; is the optimal solution at the first iteration; is the optimal solution of the i-th path point at the t-th iteration; t is the current iteration number and is an integer ≥1.

[0063] S3: Substitute the path point solution set into the improved particle swarm algorithm to determine the position of each path point; in the first iteration, the reverse roulette algorithm is used to calculate the pseudo global optimal solution from the path point solution set. The pseudo global optimal solution as the optimal solution for the first iteration.

[0064] S4: Construct a fitness function F related to the location of the path pointitness , substitute the position of each path point into the fitness function F itness Calculate the optimal solution under the current number of iterations

[0065] Specifically, step S4 includes: constructing a fitness function F related to the location of the path point itness , and use the Meltrov criterion to determine the optimal solution under the current number of iterations

[0066] S5: Repeat steps S2-S4 until the maximum number of iterations is reached and the global optimal solution X is determined. g , the global optimal solution X g The path formed by the corresponding path points serves as the penetration path of the swarm UAVs; the number of iterations is set by the user according to actual needs.

[0067] In step S2, a dynamic weight coefficient is introduced into the particle swarm algorithm. The dynamic weight coefficient is one of the important parameters for controlling the change of the path point speed, which indicates the degree to which the current path point is affected by the speed of the previous generation of path points. Dynamically changing the dynamic weight coefficient means that with continuous iterations, the value of the dynamic weight coefficient is gradually adjusted, thereby affecting the movement behavior of the path point. In this embodiment, the expression of the dynamic weight coefficient is:

[0068] ω t =ω max -0.25*log10 t

[0069] In the formula, ω t is the dynamic weight coefficient at the tth iteration; ω max is the maximum dynamic weight coefficient, which is 1; t is the current iteration number.

[0070] When the number of iterations is low, ω t The larger the value, the more path points can fully search the current space and jump out of the local optimal solution. As the number of iterations increases, ω t It gradually decreases and finally approaches the set minimum value of the dynamic weight coefficient.

[0071] In step S2, a learning coefficient variation strategy is introduced into the particle swarm algorithm, wherein the learning coefficient includes a cognitive coefficient and a social coefficient. The cognitive coefficient c 1 Represents the degree to which a path point is affected by its own experience, which determines that a path point will be adjusted according to its own historical optimal position during the search process. Social coefficient c 2Represents the degree of influence of the group on the path point, which determines that a path point will be guided by the position information of other particles during the search process. A larger social coefficient and a smaller cognitive coefficient will make the path point more likely to be influenced by the group, which is helpful for global search; conversely, it will make the particle pay more attention to the local information of the individual, which is helpful for local search. In this embodiment, the update formulas of the cognitive coefficient and the social coefficient are as follows:

[0072]

[0073] In the formula, c 1max is the maximum value of the cognitive coefficient; c 1min is the minimum value of the cognitive coefficient; t max is the maximum number of iterations, t is the current number of iterations; c 2max is the maximum value of the social coefficient; c 2min is the minimum value of the social coefficient; is the cognitive coefficient at the tth iteration; is the social coefficient at the tth iteration.

[0074] In the early stage of iteration, each path point can search in its own direction as much as possible to reduce the influence of other path points on itself. As the number of iterations increases, in the later stage of iteration, it is hoped that more path points can search towards the global optimal point to enhance the global search ability of the entire iteration process and avoid falling into the local optimal solution.

[0075] In step S2, a chaotic coefficient is introduced into the particle swarm algorithm. Chaotic distribution is a type of mathematical distribution that is disordered, unpredictable, and extremely sensitive to initial conditions. Compared with ordinary randomness, the main advantages of chaotic distribution are its higher randomness and unpredictability, as well as its ability to guide the search process. In this embodiment, the calculation formula of the chaotic coefficient is:

[0076]

[0077] In the formula, r is a constant, with a value of 0.7; a t-1 is the chaos coefficient at the (t-1)th iteration; a t is the chaos coefficient at the tth iteration.

[0078] In step S3, the roulette algorithm is a probability-based selection method, in which each selection object (or individual) is assigned an area on the roulette wheel, and its area is proportional to its probability of being selected. It is often used in optimization algorithms such as genetic algorithms and evolutionary algorithms. The roulette algorithm is improved to obtain the reverse roulette algorithm, and the probability p of selecting a path point from the path point solution set is calculated using the reverse roulette algorithm. i for:

[0079]

[0080] In the formula, F itness is the fitness function; N′ is the first 10% of the path points with lower fitness values; i is the number of path points.

[0081] In step S4, a fitness function F related to the position of the path point is constructed. itness , use the Meltrov criterion to determine the optimal solution under the current number of iterations The Meltrov criterion can introduce a certain degree of randomness in the search process, increasing the exploration ability of path points in the solution space, rather than just moving towards the direction of the global optimal solution, avoiding falling into the local optimal solution.

[0082] In step S4, the fitness function F itness for:

[0083] F itness =F 1 +F 2 +F 3 +F 4

[0084] In the formula, F 1 is the terrain threat cost function; F 2 is the energy consumption cost function; F 3 is the radar threat cost function; F 4 is the flight altitude cost function; where F 1 and F 2 They are all functions related to the position of the path points.

[0085] (1) Terrain threat cost function F 1 for:

[0086] F 1 =R des *K 1

[0087] In the formula, R des is the crash rate of drones in the swarm, which is a function related to the position of the path point; K 1 is a constant with a value of 0.3; see Figure 5 In the face of terrain changes and complex terrain, the flight path of the swarm UAV is directly affected by the terrain. The flight altitude and path planning of the formation need to be adjusted according to factors such as the undulation of the terrain and the distribution of obstacles. Set the terrain center of the swarm UAV to C k , the threat radius is R k , the distance between the formation center and the obstacle is L rd , R is the radius of the UAV formation, N i is the i-th path point of the cluster UAV, Ni+1 is the (i+1)th path point of the swarm UAV.

[0088] See also Figure 5 When the swarm drones penetrate, the ultraviolet light guidance link is used to keep the swarm drones under control. Figure 5 The topology diagram of the cluster UAV formation is shown in Figure 2. The distance between cluster UAVs is maintained at 5m, the radius of the UAV is 0.5m, the lateral width of the UAV formation is 13m, and the distances between the formation central axis and UAV 4 and UAV 2 are 6.5m and 4m respectively.

[0089] Therefore, the number of crashes N under different collision conditions des The calculation formula is:

[0090]

[0091] L rd =x d -x i

[0092] Where, L rd is the distance between the formation center and the obstacle, unit: m; x d is the x-coordinate value of the obstacle with the center of the terrain as the origin; i is the x-coordinate value of the ith path point with the center of the terrain as the coordinate origin.

[0093]

[0094] Where N des is the number of crashes under different collision conditions, unit: piece; N total is the number of drones in the formation, unit: piece.

[0095] (2) In order to make the UAV achieve efficient breakthrough, the planned path should be minimized. The energy consumption of the UAV is divided into flight energy consumption and ultraviolet communication energy consumption. The flight energy consumption is positively correlated with the path length. Therefore, the energy consumption rate is related to the energy consumption cost function F 2 for:

[0096]

[0097] Energy consumption cost function F 2 for:

[0098] F 2 =R e *K 2

[0099] In the formula, R e is the energy consumption rate, which is a function related to the position of the path point; K 2is a constant, with a value of 0.2; b 1 is the path length and energy mapping coefficient, which is generally taken as 0.5; n is the total number of path points in the path point solution set, in units of pieces; x i+1 ,y i+1 and z i+1 are the coordinates of the (i+1)th path point with the center of the terrain as the coordinate origin, in meters; x i ,y i and z i are the coordinates of the i-th path point with the center of the terrain as the coordinate origin, in meters; t f is the flight time of the swarm UAV, unit: s; E A The UV energy consumption of cluster head node data fusion; D j are the other UAVs except the leading UAV; k is the communication bit data volume per unit time; m is the total number of UAVs in the formation, unit: piece; E j is the node energy consumption of the jth UAV in the formation; E max is the maximum energy consumption of swarm drones.

[0100] (3) During the UAV formation’s penetration, the enemy’s radar system’s detection capability may make it a potential threat factor. Radar can detect the presence of UAVs at a long distance, which in turn affects the concealment and secrecy of the mission. When the formation is performing a mission with strong concealment and low altitude flight, it should avoid radar monitoring to reduce the possibility of being detected by the enemy. Among them, the radar receives the electromagnetic echo signal power p scattered by the target r for:

[0101]

[0102] Where G is the signal receiving and transmitting gain; p t is the peak transmission power; λ is the radar wavelength; σ is the radar scattering area of ​​the target; R is the distance between the target and the radar.

[0103] In the UAV penetration scenario, the radar echo power is mainly affected by the distance R between the UAV and the radar. Therefore, the single radar detection probability model is simplified to:

[0104]

[0105] In the formula, p s is the probability that the sth radar detects the UAV; R max is the maximum detection radius of the radar;

[0106] Radar threat cost function F 3 for:

[0107]

[0108] Where P det is the probability of all radars being exposed; K 3 is a constant with a value of 0.3; q is the total number of radars, unit: piece.

[0109] (4) When performing a low-altitude penetration mission, the UAV will always maintain a pre-defined low-altitude altitude range during the execution process. This altitude range is set to ensure the low-altitude nature of the mission and to meet the specific requirements of the mission. When the UAV exceeds the predetermined altitude limit during the mission execution, the system will determine that the low-altitude penetration mission is in a failed state. The flight altitude cost F in this paper is 4 for:

[0110]

[0111] Where n is the number of path points in the path point solution set, unit: piece; k 4 is a constant, with a value of 0.2; z i is the flight height of the swarm UAV at the i-th path point, in meters.

[0112] Therefore, the total cost function, that is, the fitness function F itness for:

[0113] F itness =F 1 +F 2 +F 3 +F 4

[0114] It should be noted that the particles in the particle group in this embodiment are equivalent to the path points, and the dimensions of the particles are equivalent to the positions of the path points, that is, the coordinates of the path points.

[0115] Example 2

[0116] See also Figure 2 The cluster UAV penetration path planning system of this embodiment is applied to the cluster UAV penetration path planning method of embodiment 1, and includes an acquisition module, an algorithm construction module, a path point position determination module, a calculation module and an iteration module;

[0117] Acquisition module: used to obtain the path point solution set of the cluster drones;

[0118] Algorithm building module: used to introduce dynamic weight coefficients into the particle swarm algorithm and build an improved particle swarm algorithm;

[0119] Waypoint location determination module: used to substitute the waypoint solution set into the improved particle swarm algorithm to determine the location of each waypoint;

[0120] Calculation module: used to construct the fitness function F related to the location of the path pointitness , substitute the position of each path point into the fitness function F itness Calculate the optimal solution under the current number of iterations

[0121] And iterative module: used to repeatedly iterate the algorithm construction module, the path point location determination module and the calculation module until the maximum number of iterations is reached to determine the global optimal solution X g , the global optimal solution X g The path formed by the corresponding path points serves as the penetration path of the swarm UAVs.

[0122] The swarm UAV penetration path planning system of this embodiment corresponds to the swarm UAV penetration path planning method of Embodiment 1. Specifically, for the contents of the acquisition module, algorithm construction module, path point position determination module, calculation module and iteration module, please refer to the description of the swarm UAV penetration path planning method part of Embodiment 1.

[0123] Example 3

[0124] The present embodiment provides a memory storing a program file, wherein the program file is executed to implement program instructions formed by the cluster drone penetration path planning method of embodiment 1.

[0125] The memory in this embodiment may specifically include random access memory (RAM), internal memory, read-only memory (ROM), programmable ROM, erasable programmable ROM, register, hard disk, removable disk or CD-ROM. It should be noted that those skilled in the art may specifically select the form and type of memory according to actual production and use requirements, and no further limitation is made in this embodiment.

[0126] Example 4

[0127] This embodiment is an electronic device, including a processor and a memory coupled to each other, wherein the memory is used to store program instructions formed by the cluster drone penetration path planning method of Example 1; the processor is used to execute the program instructions stored in the memory.

[0128] The electronic device in this embodiment includes any electronic device that can execute program instructions, such as a computer, a mobile terminal or a wearable device.

[0129] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some or all of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the present invention.

Claims

1. A cluster UAV penetration path planning method, characterized in that: The following steps are involved: S1: Obtain the path point solution set of the swarm drones; S2: Introduce dynamic weight coefficient into the particle swarm algorithm to construct an improved particle swarm algorithm; S3: Substitute the path point solution set into the improved particle swarm algorithm to determine the position of each path point; S4: Construct a fitness function F related to the location of the path point itness , substitute the position of each path point into the fitness function F itness Calculate the optimal solution under the current number of iterations S5: Repeat steps S2-S4 until the maximum number of iterations is reached and the global optimal solution X is determined. g , the global optimal solution X g The path formed by the corresponding path points serves as the penetration path of the swarm UAVs.

2. The method for cluster UAV penetration path planning according to claim 1 is characterized in that: In step S2, the improved particle swarm algorithm is: In the formula, is the velocity of the i-th path point at the t-th iteration; ω t is the dynamic weight coefficient at the tth iteration; is the speed of the i-th path point at the (t-1)th iteration; is the cognitive coefficient at the (t-1)th iteration; a t-1 is the chaos coefficient at the (t-1)th iteration; X ip is the optimal solution of the i-th path point in history; is the optimal solution of the i-th path point at the (t-1)th iteration; is the social coefficient at the (t-1)th iteration; is the optimal solution at the first iteration; is the optimal solution of the i-th path point at the t-th iteration; t is the current iteration number and is an integer ≥1.

3. The cluster UAV penetration path planning method according to claim 2 is characterized in that: The expression of the dynamic weight coefficient is: oh t =ω max -0.25*log10 t In the formula, ω t is the dynamic weight coefficient at the tth iteration; ω max is the maximum dynamic weight coefficient, which is 1; t is the current iteration number.

4. The cluster UAV penetration path planning method according to claim 2 is characterized in that: In step S3, in the first iteration, the pseudo global optimal solution is calculated from the path point solution set using the reverse roulette algorithm. The pseudo global optimal solution as the optimal solution for the first iteration.

5. The method for cluster UAV penetration path planning according to claim 1, characterized in that: The step S4 comprises: constructing a fitness function F related to the position of the path point itness , and use the Meltrov criterion to determine the optimal solution under the current number of iterations 6. The swarm UAV penetration path planning method according to claim 1 or 5, characterized in that: The fitness function F itness for: <h2 style=";text-align:left;direction:ltr">F<h2 style=";text-align:left;direction:ltr"> itness <h2 style=";text-align:left;direction:ltr"> (F1+F2+F3+F4) Wherein, F1 is the terrain threat cost function; F2 is the energy consumption cost function; F3 is the radar threat cost function; F4 is the flight altitude cost function; and both F1 and F2 are functions related to the position of the waypoint.

7. The method for cluster UAV penetration path planning according to claim 6 is characterized in that: The terrain threat cost function F1 is: F1=R des *K1 In the formula, R des is the crash rate of drones in the swarm, which is a function related to the position of the path point; K1 is a constant with a value of 0.3; The energy consumption cost function F2 is: F2=R e *K2 In the formula, R e is the energy consumption rate, which is a function related to the position of the path point; K2 is a constant, with a value of 0.2; The radar threat cost function F3 is: F3=P det *K3 Where P det is the probability of all radars being exposed; K3 is a constant, and its value is 0.

3.

8. A swarm UAV penetration path planning system, applied to the swarm UAV penetration path planning method according to claim 1, comprising: Acquisition module: used to obtain the path point solution set of the cluster drones; Algorithm building module: used to introduce dynamic weight coefficients into the particle swarm algorithm and build an improved particle swarm algorithm; Waypoint location determination module: used to substitute the waypoint solution set into the improved particle swarm algorithm to determine the location of each waypoint; Calculation module: used to construct the fitness function F related to the location of the path point itness , substitute the position of each path point into the fitness function F itness Calculate the optimal solution under the current number of iterations And iterative module: used to repeatedly iterate the algorithm construction module, the path point location determination module and the calculation module until the maximum number of iterations is reached to determine the global optimal solution X g , the global optimal solution X g The path formed by the corresponding path points serves as the penetration path of the swarm UAVs.

9. A memory, characterized in that: A program file is stored, and the program file is executed to implement program instructions formed by the cluster drone penetration path planning method as described in any one of claims 1-7.

10. An electronic device, characterized in that: The invention comprises a processor and a memory coupled to each other, wherein the memory is used to store program instructions formed by the cluster UAV penetration path planning method according to any one of claims 1 to 7; The processor is used to execute program instructions stored in the memory.

Images