Unmanned aerial vehicle path planning method and device, electronic equipment and readable storage medium

Abstract

This application relates to the field of path optimization technology, and in particular to a method, apparatus, electronic device, and readable storage medium for UAV path planning, comprising: generating a population; determining the first and second light intensities of each individual and partitioning each individual, and executing an iterative loop based on the partitioning results; adaptively adjusting the number of individuals in the illuminated area and the shadow area; determining the first and second child individuals of each individual in the illuminated area; determining all individuals in the current generation based on gradient information and velocity information; determining the complexity of the optimal individual in the current generation, adjusting the number of path points of all individuals in the next generation based on the complexity, and selecting the initial individuals for the next round of iteration; when the number of iterations is a preset number of rounds, determining the global optimal solution as the optimal obstacle avoidance path for the UAV, thereby improving the adaptability and search efficiency in dynamic environments, and solving the problems of imbalance between global search and local development, slow convergence speed, and premature convergence in existing algorithms in dynamic and complex environments.

Description

Technical Field

[0001] This application relates to the field of path optimization technology, and in particular to a method, apparatus, electronic device and readable storage medium for unmanned aerial vehicle (UAV) path planning. Background Technology

[0002] In recent years, with the widespread application of drones in various industries, the development of high-performance drones has become a focus of attention for many countries. Drone path planning is one of the core issues in drone mission allocation. Specifically, it refers to the ability of a drone, in a specific environment, to quickly plan an optimal route from its current location to safely reach the target area, taking into account its own performance characteristics. As modern technology becomes increasingly complex, traditional drone path planning methods can no longer meet the actual needs of drone flight missions. Developing a reliable and effective path planning method has become a bottleneck for drones to efficiently complete flight missions.

[0003] For path planning of UAVs, particle swarm optimization is often used to search for the optimal path. It can converge to the global optimum with a high probability, and has a fast calculation speed and better global search capability. However, for functions with multiple local extrema, it is easy to get trapped in local optima, resulting in premature convergence.

[0004] Therefore, the above methods are still difficult to effectively and stably achieve an adaptive balance between global exploration and local development in complex dynamic path planning scenarios, and to quickly generate high-quality feasible paths. Summary of the Invention

[0005] In view of this, embodiments of this application provide a method, apparatus, electronic device, and readable storage medium for UAV path planning, in order to solve the problems of imbalance between global search and local development, slow convergence speed, and premature convergence in existing algorithms in dynamic and complex environments.

[0006] A first aspect of this application provides a method for unmanned aerial vehicle (UAV) path planning, including: Construct a solution space for the drone obstacle avoidance path, including the start point, end point, and obstacles, and generate a population corresponding to the drone obstacle avoidance path in the solution space based on a preset chaotic mapping method; The first light intensity of each individual is determined based on the fitness value of each individual in the population. The second light intensity of each individual is determined based on the shading coefficient and fitness value of each individual. Each individual is partitioned based on the first light intensity and the second light intensity to obtain the partitioning result. An iterative loop is performed on the population based on the partitioning result. The partitioning result includes the illuminated area and the shaded area. In each iteration, the number of individuals in the illuminated and shaded areas is adaptively adjusted exponentially based on the maximum and minimum proportions of the illuminated area and the shape of the control curve. The first sub-individual of each body in the illuminated area is determined based on a random step size that follows a Lévy distribution, and the second sub-individual of each body in the illuminated area is determined based on the current optimal solution of the illuminated area. The first sub-individual of each body in the shadow region is determined based on the gradient and velocity information of each body in the shadow region, and the second sub-individual of each body in the shadow region is determined based on the current optimal solution of the shadow region. Each individual in the illuminated area, each individual in the shaded area, the first child of each individual in the illuminated area, the second child of each individual in the illuminated area, the first child of each individual in the shaded area, and the second child of each individual in the shaded area are identified as all individuals in this generation. The fitness value of all individuals in this generation is calculated. The complexity of the current generation's best individual is determined based on the first indicator function. The number of path points for all individuals in the next generation is adjusted based on the complexity of the current generation's best individual. The best individuals in the current generation whose fitness values ​​are among the preset number of individuals before sorting are selected as the initial individuals for the next iteration. The first indicator function is used to calculate the number of path segments on the path of the current generation's best individual that are within the preset threat area radius. When the number of iterations is a preset number of rounds, the global optimal solution is determined as the optimal obstacle avoidance path for the UAV.

[0007] A second aspect of this application provides a drone path planning device, comprising: The construction module is used to construct the solution space of the drone obstacle avoidance path, which includes the start point, the end point and obstacles. In the solution space, a population corresponding to the drone obstacle avoidance path is generated based on a preset chaotic mapping method. The partitioning module is used to determine the first light intensity of each individual based on the fitness value of each individual in the population, determine the second light intensity of each individual based on the shading coefficient and fitness value of each individual, partition each individual based on the first light intensity and the second light intensity to obtain the partitioning result, and perform an iterative loop on the population based on the partitioning result. The partitioning result includes the illuminated area and the shaded area. The adjustment module is used to adaptively adjust the number of individuals in the illuminated and shaded areas based on the maximum and minimum proportions of the illuminated area and the shape of the control curve in each iteration loop. The first determining module is used to determine the first sub-individual of each body in the illuminated area based on a random step size that follows a Lévy distribution, and to determine the second sub-individual of each body in the illuminated area based on the current optimal solution of the illuminated area. The second determining module is used to determine the first sub-individual of each body in the shadow area based on the gradient information and velocity information of each body in the shadow area and the current optimal solution of the shadow area, and to determine the second sub-individual of each body in the shadow area based on the current optimal solution of the shadow area. The third determination module is used to determine each individual in the illuminated area, each individual in the shaded area, the first child of each individual in the illuminated area, the second child of each individual in the illuminated area, the first child of each individual in the shaded area, and the second child of each individual in the shaded area as all individuals in this generation, and to calculate the fitness value of all individuals in this generation. The filtering module is used to determine the complexity of the current generation's best individual based on the first indicator function, adjust the number of path points of all individuals in the next generation based on the complexity of the current generation's best individual, and filter out the better individuals of the current generation whose fitness values ​​are within a preset number before sorting as the initial individuals of the next iteration. The first indicator function is used to calculate the number of path segments on the path of the current generation's best individual that are within a preset threat area radius. The output module is used to determine the global optimal solution as the optimal obstacle avoidance path for the UAV when the number of iterations is a preset number of rounds.

[0008] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0009] A fourth aspect of this application provides a readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0010] Compared with the prior art, the beneficial effects of this application embodiment are as follows: by constructing a solution space for UAV obstacle avoidance paths and using chaotic mapping to generate an initial population, the diversity of the initial population and the foundation for global search are improved; by combining fitness and occlusion coefficient to divide the illuminated and shadowed areas, the population is optimized hierarchically; during the iteration process, the number of individuals in the illuminated and shadowed areas is adaptively adjusted, balancing the algorithm's global exploration and local development capabilities; new and old sub-individuals are generated using Lévy distribution, gradient and velocity information, and the optimal solution of the region, respectively, enriching the optimization mechanism and accelerating population convergence; by merging new and old individuals and uniformly calculating the fitness value, the range of optimization candidates is expanded; by using the first indicator function to evaluate path complexity and adaptively adjusting the number of path points, the obstacle avoidance safety in complex environments and the computational efficiency in simple environments are enhanced; in each iteration, the better individuals are retained for updating and iteration, reducing invalid searches and accelerating convergence; by outputting the globally optimal obstacle avoidance path at the end of the iteration, the algorithm's global exploration and local development capabilities are enhanced, and the path planning adaptability and search efficiency of UAVs in dynamic environments are improved.

[0011] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a schematic diagram illustrating an application scenario of this application embodiment; Figure 2 This is a schematic flowchart of a phototropic growth algorithm provided in an embodiment of this application; Figure 3 This is a flowchart illustrating a UAV path planning method provided in an embodiment of this application; Figure 4 This is a flowchart illustrating another UAV path planning method provided in an embodiment of this application; Figure 5 This is a schematic diagram of a three-machine multi-obstacle path planning method provided in an embodiment of this application; Figure 6 This is a schematic diagram of a three-machine multi-obstacle path planning method from different perspectives, provided in an embodiment of this application. Figure 7 This is a schematic diagram of a three-machine multi-obstacle path planning method from a top-down perspective, provided in an embodiment of this application. Figure 8 This is a schematic diagram of the structure of a drone path planning device provided in an embodiment of this application; Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0014] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0015] The following will describe in detail, with reference to the accompanying drawings, a method and apparatus for unmanned aerial vehicle (UAV) path planning according to embodiments of this application.

[0016] Figure 1 This is a schematic diagram illustrating an application scenario according to an embodiment of this application. The application scenario may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a server 104, and a network 105.

[0017] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be hardware or software. When the first terminal device 101, the second terminal device 102, and the third terminal device 103 are hardware, they can be various electronic devices with displays and supporting communication with the server 104, including but not limited to smartphones, tablets, laptops, and desktop computers. When the first terminal device 101, the second terminal device 102, and the third terminal device 103 are software, they can be installed on the aforementioned electronic devices. The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be implemented as multiple software programs or software modules, or as a single software program or software module; this application embodiment does not impose any limitations on this. Furthermore, various applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as data processing applications, instant messaging tools, social platform software, search applications, shopping applications, etc.

[0018] Server 104 can be a server that provides various services, such as a backend server that receives requests sent by terminal devices with which it has established communication connections. This backend server can receive and analyze the requests sent by the terminal devices and generate processing results. Server 104 can be a single server, a server cluster consisting of several servers, or a cloud computing service center. This application embodiment does not limit this.

[0019] It should be noted that the server 104 can be either hardware or software. When the server 104 is hardware, it can be various electronic devices that provide various services to the first terminal device 101, the second terminal device 102, and the third terminal device 103. When the server 104 is software, it can be multiple software programs or software modules that provide various services to the first terminal device 101, the second terminal device 102, and the third terminal device 103, or it can be a single software program or software module that provides various services to the first terminal device 101, the second terminal device 102, and the third terminal device 103. This application embodiment does not impose any limitations on this.

[0020] Network 105 can be a wired network using coaxial cable, twisted pair, and fiber optic connection, or it can be a wireless network that enables interconnection of various communication devices without wiring, such as Bluetooth, Near Field Communication (NFC), and Infrared. This application embodiment does not limit this.

[0021] Users can establish a communication connection with the server 104 via the network 105 through the first terminal device 101, the second terminal device 102 and the third terminal device 103 to receive or send information, etc.

[0022] It should be noted that the specific types, quantities and combinations of the first terminal device 101, the second terminal device 102, the third terminal device 103, the server 104 and the network 105 can be adjusted according to the actual needs of the application scenario, and this application embodiment does not limit this.

[0023] It should be noted that the acquisition, storage, use, and processing of data in this application embodiment all comply with the relevant provisions of national laws and regulations.

[0024] Furthermore, it should be noted that in the embodiments of this application, certain software, components, models, and other existing solutions in the industry may be mentioned. These should be considered as exemplary, and their purpose is only to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0025] First, it should be noted that this application improves upon the existing light-growing algorithm, specifically targeting the characteristics of UAV obstacle avoidance application scenarios. The principle of the light-growing algorithm will now be explained, as follows: Figure 2 As shown.

[0026] The core of the phototropic growth algorithm (PGA) is to simulate how plant cells grow in response to light. In optimization problems, this biological mechanism is mapped into mathematical language, and the correspondence is shown in Table 1: Table 1. Comparison Relationship Table

[0027] Combining the principles in the table above and Figure 2 The specific steps of the light-growing algorithm are explained.

[0028] First, the population is initialized. At the beginning of the PGA algorithm, a random population is initialized, including... A population of cells, which can be denoted as Subsequently, the population was sorted according to fitness and divided into two subpopulations: one for light and one for shade. The light subpopulation consisted of cells in the light subpopulation (…). ) represents the currently found better solution, while the cells in the shaded area ( () represents a poor solution that requires further exploration.

[0029] Among them, the number of cells in the illuminated area The ratio is determined by a random proportion to ensure the flexibility of the algorithm; the specific formula is as follows: ; in, and It is a preset proportionality constant. It is a random number between [0,1], and the number of cells in the shaded area is... .

[0030] After dividing the cells into illuminated and shaded areas, the state of the cells in the illuminated area is updated. For cells in the illuminated area (excellent), their division is aimed at local search to further explore the optimal solution.

[0031] First, the illuminated sub-cell 1 is generated using a mutation operator. This process introduces randomness and a mechanism for learning from the optimal solution, as shown in the following formula: ; in, Let i represent the current mother cell, and let i represent the i-th individual in the t-th generation population. This indicates the newly generated daughter cells. This represents the current globally optimal solution. This represents randomly selected cells, used to increase diversity. This represents a random mutation factor, taking values ​​from {-1, 1}, which controls the direction of mutation. This represents the growth limiting factor, simulating the phenomenon of biological growth rate decaying over time. The formula is: ( It is the current iteration number). Where T represents the maximum number of iterations of the algorithm, initially... Large, which is beneficial for overall exploration; later Small size allows for detailed, localized development.

[0032] Secondly, daughter cells 2 in the illuminated area are generated through an auxin redistribution operator. This process simulates the characteristic of auxin guiding cells to grow towards light, as shown in the following formula: ; This formula directly drives the cell toward the current optimal solution. Proximity is key to the convergence of the algorithm.

[0033] The goal of the state update process in the shaded region is to explore the unknown area and avoid getting trapped in local optima. Therefore, this process usually searches over a wider range to enhance its randomness.

[0034] First, the child cell 1 of the shaded region is generated using the mutation operator, as shown in the following formula: ; in, , This indicates randomly selected cells, used to increase diversity.

[0035] Then, daughter cells 2 of the shaded area are generated using the auxin redistribution operator, as shown in the following formula: ,in, This represents the optimal solution for the shaded area.

[0036] After this round of division, the newly generated daughter cells will merge with the parent cells, and then be selected based on fitness to eliminate inferior individuals and maintain the population size.

[0037] Based on the aforementioned light-growing algorithm, this application proposes a UAV path planning algorithm, such as... Figure 3 This is a flowchart illustrating a drone path planning method provided in an embodiment of this application. Figure 3 The drone path planning method can be derived from Figure 1 The terminal device executes the command. For example... Figure 3 As shown, the UAV path planning method includes: Step S301: Construct a solution space for the drone obstacle avoidance path, including the start point, end point, and obstacles. Generate a population corresponding to the drone obstacle avoidance path in the solution space based on a preset chaotic mapping method.

[0038] In some embodiments of this application, a population corresponding to the UAV obstacle avoidance path is generated in the solution space based on a preset chaotic mapping method, including: Chaotic variables for each entity in different dimensions are generated based on a preset chaotic mapping method; Based on the chaotic variables of each individual in different dimensions and the upper and lower bounds of the solution space, a population corresponding to the obstacle avoidance path of the UAV is generated.

[0039] Specifically, in the scenario of UAV obstacle avoidance path planning, a solution space corresponding to the path planning space is constructed. Each feasible solution in the solution space corresponds to a potential UAV obstacle avoidance path. Each solution is defined as an individual in the algorithm, and all individuals together constitute a population. The population size is set according to the complexity of the actual path planning. For example, it can be set to 50 to ensure the diversity of the population to cover a wider range of solution spaces, while avoiding computational redundancy caused by excessive size.

[0040] Furthermore, since the PGA algorithm uses random initialization, the initial population may be unevenly distributed, with some regions being overly dense while others remain unexplored, affecting the convergence speed and the quality of the final solution. Therefore, this application introduces the Tent chaotic mapping to generate the initial population, using the ergodicity and randomness of chaos to ensure that the initial solution is evenly distributed in the solution space.

[0041] The Tent chaotic mapping is specifically implemented through the following formula: ; in, Let n be the chaotic sequence quantity. It represents the chaotic sequence quantity of the (n+1)th iteration.

[0042] Furthermore, the chaotic sequence is mapped to the decision variable space, as shown in the following formula: ; in, This represents the initial decision variable value of the i-th individual in the j-th dimension, which is the initial position of the j-th dimension of the i-th solution in the algorithm population. It is the first The individual Chaotic quantities of dimension They are the first The lower and upper bounds of a dimension.

[0043] Step S302: Determine the first light intensity of each individual based on the fitness value of each individual in the population; determine the second light intensity of each individual based on the shading coefficient and fitness value of each individual; partition each individual based on the first light intensity and the second light intensity to obtain the partitioning result; perform an iterative loop on the population based on the partitioning result; the partitioning result includes the illuminated area and the shaded area.

[0044] In some embodiments of this application, a first light intensity for each individual is determined based on its fitness value; a second light intensity for each individual is determined based on its occlusion coefficient and fitness value; each individual is partitioned based on the first and second light intensities to obtain partitioning results; and an iterative loop is performed on the population based on the partitioning results, including: The first light intensity of each individual is determined based on the fitness value of each individual in the population. Obtain the path segment for each individual, and based on the path segment for each individual, obtain the shortest distance from each path segment to the center point of the obstacle, the distance from the start point to the end point of each individual's path segment, and a second indicator function. The second indicator function is used to determine the number of path segments for each individual whose distance from the center point of the obstacle is less than the safe distance. The occlusion coefficient of each object is determined based on the shortest distance from each path segment to the center point of the obstacle, the distance from the starting point to the ending point of each path segment, and the second indicator function. The second light intensity of each individual is determined based on its occlusion coefficient and fitness value. Individuals with the same first light intensity as the second light intensity are assigned to the illuminated area, while individuals with different first light intensities are assigned to the shaded area. The population is then iterated.

[0045] Specifically, obstacle occlusion factors are introduced into the original light intensity model to adapt to the needs of the UAV path planning scenario in this application. By introducing the obstacle factor, the rationality of the partitioning can be ensured, and the defects caused by the random partitioning of the existing technology can be reduced.

[0046] First, define the population. and the population They are divided into two groups: the illuminated area and the shaded area. The illuminated area refers to the region where high-quality solutions or better individuals are located, while the shaded area is the region where low-quality solutions are located.

[0047] Due to light intensity It is usually negatively correlated with the fitness value (the smaller the fitness, the greater the light intensity), so the first light intensity can be obtained, and the specific formula for the first light intensity is as follows: ;in, This represents the i-th path in the optimization algorithm population. It is a very small positive number. Representing a path The fitness function value.

[0048] For path Where P represents a path segment, the occlusion coefficient is defined in this application as shown in the following formula: ; Where D represents the path The total number of waypoints included, where K represents the total number of obstacles in the environment. For path The occlusion coefficient, Represents path segment , Represents path segment To the center of the obstacle The shortest distance. Represents path segment starting point To the end of the path Path distance, Indicates the second indicator function, path segment arrive If the distance is less than the safe distance, output 1; otherwise, output 0.

[0049] Therefore, after introducing the above occlusion coefficient, we can obtain the corrected light intensity formula, which is the formula for the second light intensity: ;in, It represents a very small positive number. Represents the i-th path The light intensity value.

[0050] After obtaining the first and second light intensities, the individuals in the population are adjusted based on these intensities. When the first and second light intensities of the same individual are equal, it indicates that the individual is a superior individual and is assigned to the illuminated area. When the first and second light intensities are not equal, it indicates that the individual has a long path or is close to an obstacle, belonging to a poor solution, and should be assigned to the shaded area. The above processing constitutes the initial partitioning. After the initial partitioning is completed, the iterative loop officially begins. By introducing an obstacle occlusion coefficient, the original first light intensity is corrected to obtain the second light intensity. The first and second light intensities are then compared. Compared to the random partitioning of existing technologies, partitioning guided by the first and second light intensities ensures the rationality of the partitioning.

[0051] Step S303: In each iteration, the number of individuals in the illuminated area and the shadow area is adaptively adjusted based on the maximum and minimum proportions of the illuminated area and the exponential shape of the control curve.

[0052] In some embodiments of this application, in each iteration, the number of individuals in the illuminated and shaded areas is adaptively adjusted based on the maximum and minimum proportions of the illuminated area and the exponential shape of the control curve, including: The contemporary illumination area proportion is determined based on the maximum and minimum proportions of the illumination area and the index of the control curve shape. The number of individuals in the illuminated area is determined based on the contemporary proportion of illuminated areas, and the number of individuals in the shaded area is determined based on the total number of individuals in the population and the number of individuals in the illuminated area.

[0053] Specifically, since the ratio of the illuminated area to the shadow area in the original light growth algorithm is fixed, it cannot meet the needs of different evolutionary stages. In the early stage, more exploration is needed, so the ratio of the shadow area needs to be expanded. In the later stage, more development is needed, that is, the ratio of the illuminated area needs to be expanded. Therefore, at the beginning of each iteration, the embodiments of this application adjust the partition ratio according to the number of iterations to meet the needs of different evolutionary stages and achieve a dynamic balance between exploration and development.

[0054] Specifically, the partition ratio is shown in the following formula: ; It is the proportion of the illuminated area in generation t. Set as the initial proportion of individuals in the illuminated area. The ratio of individuals in the maximum illuminated area is set. t represents the maximum number of iterations, and t represents the number of iterations in this generation.

[0055] Number of individuals in the illuminated area: ; Number of individuals in the shaded area: ; Where N represents the total number of individuals in the algorithm's population. It is the proportion of the illuminated area in the tth generation. It is the minimum and maximum ratio. It is an index that controls the shape of the curve. It is a convex function, and its growth is slow in the early stages. It is a concave function and grows rapidly in the early stages.

[0056] Step S304: Determine the first sub-individual of each body in the illuminated area based on a random step size that follows a Lévy distribution, and determine the second sub-individual of each body in the illuminated area based on the current optimal solution of the illuminated area.

[0057] In some embodiments of this application, determining the first sub-individual of each body in the illuminated area based on a random step size following a Lévy distribution, and determining the second sub-individual of each body in the illuminated area based on the current optimal solution of the illuminated area, includes: Generate a random step size that follows a Lévy distribution, and determine the first sub-individual of each individual in the illuminated area based on the random step size that follows a Lévy distribution, the current optimal solution of the illuminated area, and the random individuals in the illuminated area. The second sub-individual of each individual in the illuminated area is determined based on the current optimal solution of the illuminated area and the individual individuals in the illuminated area.

[0058] Specifically, individuals in the illuminated area require two processing steps. The first corresponds to the processing of child cell 1 in the original light-growing algorithm, i.e., generation through the mutation operator. However, since the growth factor in this method is... The decision is made. As iterations proceed, the step size approaches zero. If the algorithm gets stuck in a local region surrounded by obstacles in the early stages, it will be unable to escape in later stages. Therefore, this application uses a growth factor... Replace it with a random step size that follows a Lévy distribution. Lévy flight is characterized by a "heavy tail" distribution, which can produce large step jumps with a certain probability. The specific method is shown in the following formula: ; in and Let be a random variable that follows a normal distribution, and let λ represent the exponential parameter of the Lévy distribution. The specific formula is as follows: ; in, and In Specifically, it is determined by the following formula: ;in, and It is the standard deviation of the normal distribution; Therefore, after introducing Levy flight, the first sub-individual of each body in the illuminated zone is specifically determined by the following formula: ; When the Lévy step size is large, the cell can jump out of the current local optimum region.

[0059] The second processing method is generated through an auxin redistribution operator, specifically based on the current optimal solution of the illuminated area and the determination of each volume in the illuminated area. The processing method is shown in the following formula: ; in, This represents the optimal solution for the illuminated area.

[0060] After the above processing, the individuals in the illuminated area are obtained as the first child individuals and the second child individuals of each individual in the illuminated area.

[0061] In the process of processing the illuminated area, a random step size of Levy flight is introduced. Based on the characteristic that Levy flight can generate large step jumps with a certain probability, it can effectively reduce the situation where the original light growth algorithm gets stuck in the local area surrounded by obstacles in the early stage and cannot jump out in the later stage.

[0062] Step S305: Determine the first sub-individual of each body in the shadow area based on the gradient and velocity information of each body in the shadow area, and determine the second sub-individual of each body in the shadow area based on the current optimal solution of the shadow area.

[0063] In some embodiments of this application, the first sub-individual of each body in the shadow region is determined based on the gradient and velocity information of each body in the shadow region, and the second sub-individual of each body in the shadow region is determined based on the current optimal solution of the shadow region, including: Obtain the historical velocity information of each individual in the previous generation of the shadow region, and determine the velocity information of each individual in the shadow region based on the gradient information of the historical velocity information of each individual in the previous generation of the shadow region and the fitness value of each individual in the shadow region. The first sub-individual of each body in the shadow region is determined based on the velocity information of each body in the shadow region and the current optimal solution of the shadow region. The second sub-individual of each individual in the shadow region is determined based on the current optimal solution of the shadow region and the individual individuals in the shadow region.

[0064] Specifically, similar to the processing of the illuminated area, the processing of the shadow area also involves two methods. One corresponds to the generation of child cells 1 through the mutation operator in the original light growth algorithm. However, because the update of shadow area cells in the PGA algorithm depends on random perturbations, convergence is slow. This application draws on the momentum concept of the Adam optimizer in deep learning to introduce velocity information for each cell in the shadow area. This allows it to use historical speed information to accelerate its movement towards the illuminated area.

[0065] The formula for speed information is as follows: ; in, For the speed information of the previous generation, This refers to the velocity information of this generation, including the initial velocity information. .

[0066] in, The momentum decay coefficient (usually) ), As a growth limiting factor, For fitness exist Gradient information at the location.

[0067] By combining the PGA's shaded area update, the random difference term in the original formula is removed. Replace with speed information Afterwards, we can obtain: ; in, It is a random variable factor, because It stores the direction and magnitude of movement from the previous generation, so by introducing gradient and velocity information, it can play a role in smoothing updates and accelerating escape from inferior regions.

[0068] The second method involves generating the second sub-individual of the shaded area through the auxin redistribution operator. Specifically, this is determined based on the current optimal solution of the shaded area and the individual individuals within the shaded area, as shown in the following formula: ;in, This represents the optimal solution for the shaded area.

[0069] After the above processing, the individuals in the shaded area are obtained as the first child individuals of each individual in the shaded area and the second child individuals of each individual in the shaded area.

[0070] Step S306: Determine each individual in the illuminated area, each individual in the shaded area, the first child of each individual in the illuminated area, the second child of each individual in the illuminated area, the first child of each individual in the shaded area, and the second child of each individual in the shaded area as all individuals in this generation, and calculate the fitness value of all individuals in this generation.

[0071] Specifically, at this point, we can obtain the offspring and parents of individuals in the shaded and illuminated areas. The parents are all individuals in the illuminated and shaded areas, and the offspring are the first child of each individual in the illuminated area, the second child of each individual in the illuminated area, the first child of each individual in the shaded area, and the second child of each individual in the shaded area. The offspring and parents are then considered as all individuals in this generation and sorted according to their fitness values.

[0072] Step S307: Determine the complexity of the current generation's best individual based on the first indicator function, adjust the number of path points for all individuals in the next generation based on the complexity of the current generation's best individual, and select the better individuals of the current generation whose fitness values ​​are within a preset number before sorting as the initial individuals for the next iteration. The first indicator function is used to calculate the number of path segments on the path of the current generation's best individual that are within a preset threat area radius.

[0073] In some embodiments of this application, the complexity of the current generation's best individual is determined based on a first indicator function; the number of path points for all individuals in the next generation is adjusted based on the complexity of the current generation's best individual; and the best individuals of the current generation whose fitness values ​​are among the preset number before sorting are selected as the initial individuals for the next iteration, including: Determine the total number of obstacles in the solution space and the number of path points corresponding to the best individual in this generation; The complexity of the current generation's best individual is determined based on the total number of obstacles in the solution space, the number of path points of the current generation's best individual, and the first indicator function. The first indicator function is used to calculate the number of path segments in the current generation whose distance from the obstacle is less than the threat distance. The threat distance is the radius of the preset threat area corresponding to each obstacle. The number of path points for the next generation of individuals is adjusted based on the complexity, minimum number of path points, maximum number of path points, and baseline number of path points of the current generation's best individual. All individuals in this generation are sorted based on their fitness values. A predetermined number of the best individuals in this generation that are at the top of the sort order are selected as the initial individuals for the next iteration.

[0074] In this embodiment, a fixed number of path points in the PGA algorithm path planning cannot adapt to environments of varying complexity. Simple regions require fewer path points to improve efficiency, while complex regions require more path points for precise obstacle avoidance. Therefore, this application designs an adaptive adjustment mechanism for the number of path points, dynamically adding or deleting path points based on environmental complexity.

[0075] First, determine the complexity of the best individual in this generation. The complexity assessment is based on the following formula: ; in, Represents the best individual in this generation. The complexity is [0,1]. Indicates the number of path points. Indicates the total number of obstacles. Represents path segment To the The minimum distance to each obstacle Indicates the first The radius of the threat zone of an obstacle. This indicates a safety margin, i.e., an additional buffer distance. It is the first indicator function, used to count the number of path segments close to obstacles on the path.

[0076] After determining the complexity, the number of path points for each entity in this generation is adjusted based on the minimum number of path points, the maximum number of path points, and the baseline number of path points. The adjustment is based on the following formula: ; in, This indicates the number of path points after adjustment. This represents the minimum number of path points. Indicates the maximum number of path points. Indicates the number of baseline path points. This represents an adjustment coefficient used to control the degree of influence of complexity. Let C(X) be the floor function, and let C(X) be the normalized index of path complexity.

[0077] By adjusting the number of path points for the next generation of individuals based on the optimal individual complexity, different obstacle avoidance scenarios can be effectively adapted. Since simple areas require fewer path points to improve efficiency, while complex areas require more path points for accurate obstacle avoidance, the introduction of the process of adaptively adjusting the number of path points based on complexity can take into account both scenarios with many obstacles and scenarios with few obstacles, thus improving the applicability of the algorithm.

[0078] Step S308: When the number of iterations is a preset number of rounds, the global optimal solution is determined as the optimal obstacle avoidance path for the UAV.

[0079] Furthermore, the fitness values ​​in this application will be explained.

[0080] Define a comprehensive fitness value To evaluate a path The fitness value is typically composed of multiple sub-objectives, indicating their strengths and weaknesses. ; in, For path length cost, Cost of threats / obstacles For flight altitude costs, To smooth out costs, , , , These are the weighting coefficients corresponding to each cost item.

[0081] Therefore, it is necessary to determine the final fitness value by identifying the above four cost information points. First, the path length cost is explained, and its determination formula is as follows: ; Secondly, the cost of acquiring threats / obstacles, for the first The cost and distance of each obstacle and path segment. It is inversely proportional, therefore, its formula is as follows: ; in, ; f represents the piecewise cost function, δ represents the safety buffer distance, and when considering the cost of flight altitude, the th step of path X needs to be considered. Flight altitude Limited to Therefore, its specific formula is as follows: ; ; This represents the height penalty cost for a single path segment. For smoothing costs, it is necessary to calculate the turning angle. To assess the feasibility of the path and avoid severe jitter. To satisfy kinematic constraints and penalize excessive turning angles, the specific formula is as follows: ; ; in, , This represents the initial total number of waypoints, and D represents the adjusted total number of waypoints. Indicates the first The coordinates of the path points It is the first Path points The cosine value of the turning angle, and This indicates the coordinates of the path points adjacent to it.

[0082] Find the optimal path Make: ,in, The solution space represents the path of the drone. Represents the set of real numbers. The path X of the UAV is a vector containing three-dimensional coordinate points D, and all possible paths belong to the real solution space of this 3D dimension.

[0083] In summary, as Figure 4As shown, this application, based on the original light-growing algorithm and combined with the characteristics of UAV obstacle avoidance scenarios, first initializes the population through chaotic mapping to ensure the uniform distribution of individuals in the population, and further adjusts the lighting model to incorporate obstacle occlusion coefficients to ensure more reasonable partitioning. At the beginning of each loop, adaptive dynamic partitioning is used to achieve a dynamic balance between exploration and development. In the processing of individuals in the illuminated area, Levi flight is introduced to reduce the defect of being trapped in local areas surrounded by obstacles in the early stage and unable to escape in the later stage. In the shadow area, velocity information is introduced for each cell in the shadow area, enabling it to accelerate its movement towards the illuminated area using historical velocity and gradient information. The number of path points for the next generation of individuals is adaptively adjusted to effectively adapt to environments of different complexities. Based on the above steps and technical processing, the UAV can adapt to scenarios with different complex obstacles and generate accurate and excellent path planning results.

[0084] This application conducts path planning simulation experiments in the Matlab R2022 environment. To ensure that the UAV can effectively avoid obstacles and safely reach the target point, the threat area is modeled. The experiment sets up maps with multiple UAVs under multiple threat states to simulate the real-time obstacle avoidance path generation of highly complex UAV flights. Map range The parameters of the obstacle threat model are shown in Table 2.

[0085] Table 2 Barrier Threat Model Parameters

[0086] The number of drones is set to 3, with starting coordinates of (10, 10, 8), (10, 20, 10), and (10, 30, 12) and ending coordinates of (90, 90, 15), (90, 80, 18), and (90, 70, 20). The simulation process and results are as follows.

[0087] First, the process of growing towards light begins, which involves a broad search: Algebra 30 / 200, optimal fitness = 84.88, light intensity = 337.53, collision-free plants = 50 / 50; Algebra 60 / 200, optimal fitness = 62.07, light intensity = 332.39, collision-free plants = 50 / 50; Algebra 90 / 200, optimal fitness = 62.07, light intensity = 332.39, collision-free plants = 50 / 50; Algebra 120 / 200, optimal fitness = 62.07, light intensity = 332.39, collision-free plants = 50 / 50; Algebra 150 / 200, optimal fitness = 42.75, light intensity = 320.38, collision-free plants = 50 / 50; Algebra 180 / 200, optimal fitness = 15.31, light intensity = 341.93, collision-free plants = 50 / 50; Algebra 200 / 200, optimal fitness = 1.29, light intensity = 310.24, collision-free plants = 50 / 50; After phototropic growth is completed, the final optimal fitness is 1.29; the final number of collision-free plants is 50 / 50. Final path safety distance check process: Drone 1 - minimum distance: 9.01m; Drone 2 - minimum distance: 9.22m; Drone 3 - Minimum distance: 7.97m; The final result showed that all paths were safe, and the drone successfully found the shortest path to avoid the obstacles.

[0088] Therefore, combining Figure 5 , Figure 6 and Figure 7 As shown, the three drones effectively achieved drone path planning in multi-obstacle scenarios through extensive obstacle avoidance path search and the improved light-growing algorithm.

[0089] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0090] Figure 8 This is a schematic diagram of a drone path planning device provided in an embodiment of this application. Figure 8 As shown, the UAV path planning device includes: The construction module 810 is used to construct the solution space of the UAV obstacle avoidance path, including the start point, the end point and obstacles, and generate the population corresponding to the UAV obstacle avoidance path in the solution space based on a preset chaotic mapping method. The partitioning module 820 is used to determine the first light intensity of each individual based on the fitness value of each individual in the population, determine the second light intensity of each individual based on the shading coefficient and fitness value of each individual, partition each individual based on the first light intensity and the second light intensity to obtain the partitioning result, and perform an iterative loop on the population based on the partitioning result. The partitioning result includes the illuminated area and the shaded area. The adjustment module 830 is used to adaptively adjust the number of individuals in the illuminated area and the shadow area based on the maximum and minimum proportions of the illuminated area and the shape of the control curve in each iteration loop. The first determining module 840 is used to determine the first sub-individual of each body in the illuminated area based on a random step size that follows a Lévy distribution, and to determine the second sub-individual of each body in the illuminated area based on the current optimal solution of the illuminated area. The second determining module 850 is used to determine the first sub-individual of each body in the shadow area based on the gradient information and velocity information of each body in the shadow area, and to determine the second sub-individual of each body in the shadow area based on the current optimal solution of the shadow area. The third determination module 860 is used to determine each individual in the illuminated area, each individual in the shaded area, the first child of each individual in the illuminated area, the second child of each individual in the illuminated area, the first child of each individual in the shaded area, and the second child of each individual in the shaded area as all individuals in this generation, and to calculate the fitness value of all individuals in this generation. The filtering module 870 is used to determine the complexity of the current generation's best individual based on the first indicator function, adjust the number of path points of all individuals in the next generation based on the complexity of the current generation's best individual, and filter out the better individuals of the current generation whose fitness values ​​are within a preset number before sorting as the initial individuals of the next iteration. The first indicator function is used to calculate the number of path segments on the path of the current generation's best individual that are within a preset threat area radius. Output module 880 is used to determine the global optimal solution as the optimal obstacle avoidance path for the UAV when the number of iterations is a preset number of rounds.

[0091] In some embodiments of this application, the construction module 810 is specifically used for: Chaotic variables for each entity in different dimensions are generated based on a preset chaotic mapping method; Based on the chaotic variables of each individual in different dimensions and the upper and lower bounds of the solution space, a population corresponding to the obstacle avoidance path of the UAV is generated.

[0092] In some embodiments of this application, the partitioning module 820 is specifically used for: The first light intensity of each individual is determined based on the fitness value of each individual in the population. Obtain the path segment for each individual, and based on the path segment for each individual, obtain the shortest distance from each path segment to the center point of the obstacle, the distance from the start point to the end point of each individual's path segment, and a second indicator function. The second indicator function is used to calculate the number of path segments for each individual whose shortest distance to the center point of the obstacle is less than the safe distance. The occlusion coefficient of each object is determined based on the shortest distance from each path segment to the center point of the obstacle, the distance from the start point of the path segment to the end point of the path, and the second indicator function. The second light intensity of each individual is determined based on its occlusion coefficient and fitness value. Individuals with the same first light intensity as the second light intensity are assigned to the illuminated area, while individuals with different first light intensities are assigned to the shaded area. The population is then iterated.

[0093] In some embodiments of this application, the adjustment module 830 is specifically used for: The contemporary illumination area proportion is determined based on the maximum and minimum proportions of the illumination area and the index of the control curve shape. The number of individuals in the illuminated area is determined based on the contemporary proportion of illuminated areas, and the number of individuals in the shaded area is determined based on the total number of individuals in the population and the number of individuals in the illuminated area.

[0094] In some embodiments of this application, the first determining module 840 is specifically used for: Generate a random step size that follows a Lévy distribution, and determine the first sub-individual of each individual in the illuminated area based on the random step size that follows a Lévy distribution, the current optimal solution of the illuminated area, and the random individuals in the illuminated area. The second sub-individual of each individual in the illuminated area is determined based on the current optimal solution of the illuminated area and the individual individuals in the illuminated area.

[0095] In some embodiments of this application, the second determining module 850 is specifically used for: Obtain the historical velocity information of each individual in the previous generation of the shadow region, and determine the velocity information of each individual in the shadow region based on the gradient information of the historical velocity information of each individual in the previous generation of the shadow region and the fitness value of each individual in the shadow region. The first sub-individual of each body in the shadow region is determined based on the velocity information of each body in the shadow region and the current optimal solution of the shadow region. The second sub-individual of each individual in the shadow region is determined based on the current optimal solution of the shadow region and the individual individuals in the shadow region.

[0096] In some embodiments of this application, the screening module 870 is specifically used for: Determine the total number of obstacles in the solution space and the number of path points corresponding to the best individual in this generation; The complexity of the current generation's best individual is determined based on the total number of obstacles in the solution space, the number of path points of the current generation's best individual, and the first indicator function. The first indicator function is used to calculate the number of path segments in the current generation whose distance from the obstacle is less than the threat distance. The threat distance is the radius of the preset threat area corresponding to each obstacle. The number of path points for the next generation of individuals is adjusted based on the complexity, minimum number of path points, maximum number of path points, and baseline number of path points of the current generation's best individual. All individuals in this generation are sorted based on their fitness values. A predetermined number of the best individuals in this generation that are at the top of the sort order are selected as the initial individuals for the next iteration.

[0097] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0098] Figure 9 This is a schematic diagram of the electronic device 9 provided in an embodiment of this application. Figure 9 As shown, the electronic device 9 of this embodiment includes a processor 901, a memory 902, and a computer program 903 stored in the memory 902 and executable on the processor 901. When the processor 901 executes the computer program 903, it implements the steps in the various method embodiments described above. Alternatively, when the processor 901 executes the computer program 903, it implements the functions of each module / unit in the various device embodiments described above.

[0099] Electronic device 9 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 9 may include, but is not limited to, processor 901 and memory 902. Those skilled in the art will understand that... Figure 8 This is merely an example of electronic device 9 and does not constitute a limitation on electronic device 9. It may include more or fewer components than shown, or different components.

[0100] The processor 901 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0101] The memory 902 can be an internal storage unit of the electronic device 9, such as a hard disk or RAM of the electronic device 9. The memory 902 can also be an external storage device of the electronic device 9, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., equipped on the electronic device 9. The memory 902 can also include both internal and external storage units of the electronic device 9. The memory 902 is used to store computer programs and other programs and data required by the electronic device.

[0102] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0103] If integrated modules / units are implemented as software functional units and sold or used as independent products, they can be stored in a readable storage medium (e.g., a computer-readable storage medium). Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which may be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable storage medium may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0104] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for unmanned aerial vehicle path planning, characterized in that, include: Construct a solution space for the drone obstacle avoidance path, including the start point, end point, and obstacles, and generate a population corresponding to the drone obstacle avoidance path in the solution space based on a preset chaotic mapping method. The first light intensity of each individual is determined based on the fitness value of each individual in the population. The second light intensity of each individual is determined based on the shading coefficient and fitness value of each individual. Each individual is partitioned based on the first light intensity and the second light intensity to obtain the partitioning result. An iterative loop is performed on the population based on the partitioning result. The partitioning result includes illuminated areas and shaded areas. In each iteration, the number of individuals in the illuminated and shaded areas is adaptively adjusted exponentially based on the maximum and minimum proportions of the illuminated area and the shape of the control curve. The first sub-individual of each body in the illuminated area is determined based on a random step size that follows a Lévy distribution, and the second sub-individual of each body in the illuminated area is determined based on the current optimal solution of the illuminated area. The first sub-individual of each body in the shadow region is determined based on the gradient and velocity information of each body in the shadow region, and the second sub-individual of each body in the shadow region is determined based on the current optimal solution of the shadow region. Each individual in the illuminated area, each individual in the shaded area, the first child of each individual in the illuminated area, the second child of each individual in the illuminated area, the first child of each individual in the shaded area, and the second child of each individual in the shaded area are identified as all individuals in this generation. The fitness value of all individuals in this generation is calculated. The complexity of the current generation's best individual is determined based on the first indicator function. The number of path points of all individuals in the next generation is adjusted based on the complexity of the current generation's best individual. The better individuals of the current generation whose fitness values ​​are among the preset number of individuals before sorting are selected as the initial individuals for the next iteration. The first indicator function is used to calculate the number of path segments on the path of the current generation's best individual that are within the preset threat area radius. When the number of iterations is a preset number of rounds, the global optimal solution is determined as the optimal obstacle avoidance path for the UAV.

2. The method of claim 1, wherein, The construction includes a solution space for the drone obstacle avoidance path, which includes a start point, an end point, and obstacles. Within this solution space, a population corresponding to the drone obstacle avoidance path is generated based on a preset chaotic mapping method, including: Chaotic variables for each entity in different dimensions are generated based on a preset chaotic mapping method; Based on the chaotic variables of each individual in different dimensions and the upper and lower bounds of the solution space, a population corresponding to the obstacle avoidance path of the UAV is generated.

3. The method of claim 1, wherein, The process involves determining a first light intensity for each individual based on its fitness value, determining a second light intensity for each individual based on its occlusion coefficient and fitness value, partitioning each individual into regions based on the first and second light intensities, obtaining partitioning results, and performing an iterative loop on the population based on the partitioning results, including: The first light intensity of each individual is determined based on the fitness value of each individual in the population. Obtain the path segment for each individual, and based on the path segment for each individual, obtain the shortest distance from each path segment to the center point of the obstacle, the distance from the start point to the end point of each individual's path segment, and a second indicator function. The second indicator function is used to calculate the number of path segments for each individual whose shortest distance to the center point of the obstacle is less than the safe distance. The occlusion coefficient of each body is determined based on the shortest distance from each path segment to the center point of the obstacle, the distance from the start point of the path segment to the end point of the path, and the second indicator function. The second light intensity of each individual is determined based on its occlusion coefficient and fitness value. Individuals with equal first and second light intensities are assigned to the illuminated area, while individuals with unequal first and second light intensities are assigned to the shaded area. The population is then iterated.

4. The method of claim 1, wherein, In each iteration, the number of individuals in the illuminated and shaded areas is adaptively adjusted exponentially based on the maximum and minimum proportions of the illuminated area and the shape of the control curve, including: The contemporary illumination area proportion is determined based on the maximum and minimum proportions of the illumination area and the index of the control curve shape. The number of individuals in the illuminated area is determined based on the contemporary proportion of illuminated areas, and the number of individuals in the shaded area is determined based on the total number of individuals in the population and the number of individuals in the illuminated area.

5. The method of claim 1, wherein, The process of determining the first sub-individual of each body in the illuminated area based on a random step size following a Lévy distribution, and determining the second sub-individual of each body in the illuminated area based on the current optimal solution of the illuminated area, includes: Generate a random step size that follows a Lévy distribution, and determine the first sub-individual of each individual in the illuminated area based on the random step size that follows a Lévy distribution, the current optimal solution of the illuminated area, and the random individuals in the illuminated area. The second sub-individual of each individual in the illuminated area is determined based on the current optimal solution of the illuminated area and the individual individuals in the illuminated area.

6. The method of claim 1, wherein, The process of determining the first sub-individual of each body in the shadow region based on the gradient and velocity information of each body in the shadow region, and determining the second sub-individual of each body in the shadow region based on the current optimal solution of the shadow region, includes: Obtain the historical velocity information of each individual in the previous generation of the shadow region, and determine the velocity information of each individual in the shadow region based on the gradient information of the historical velocity information of each individual in the previous generation of the shadow region and the fitness value of each individual in the shadow region. The first sub-individual of each body in the shadow region is determined based on the velocity and gradient information of each body in the shadow region and the current optimal solution of the shadow region. The second sub-individual of each individual in the shadow region is determined based on the current optimal solution of the shadow region and the individual individuals in the shadow region.

7. The method of claim 1, wherein, The process of determining the complexity of the current generation's best individual based on the first indicator function, and selecting the better individuals from the current generation whose fitness values ​​are among the preset number before sorting, as the initial individuals for the next iteration, includes: Determine the total number of obstacles in the solution space and the number of path points corresponding to the best individual in this generation; The complexity of the current generation's best individual is determined based on the total number of obstacles in the solution space, the number of path points of the current generation's best individual, and the first indicator function. The first indicator function is used to calculate the number of path segments in the current generation's best individual whose distance from the obstacle is less than the threat distance. The threat distance is the radius of the preset threat area corresponding to each obstacle. The number of path points for the next generation of individuals is adjusted based on the complexity, minimum number of path points, maximum number of path points, and baseline number of path points of the current generation's best individual. All individuals in this generation are sorted based on their fitness values. A predetermined number of the best individuals in this generation that are at the top of the sort order are selected as the initial individuals for the next iteration.

8. An unmanned aerial vehicle path planning apparatus, comprising: include: The construction module is used to construct the solution space of the UAV obstacle avoidance path, which includes the start point, the end point and obstacles, and generate the population corresponding to the UAV obstacle avoidance path in the solution space based on a preset chaotic mapping method. The partitioning module is used to determine the first light intensity of each individual based on the fitness value of each individual in the population, determine the second light intensity of each individual based on the occlusion coefficient and fitness value of each individual, partition each individual based on the first light intensity and the second light intensity to obtain the partitioning result, and perform an iterative loop on the population based on the partitioning result. The partitioning result includes illuminated areas and shaded areas. The adjustment module is used to adaptively adjust the number of individuals in the illuminated and shaded areas based on the maximum and minimum proportions of the illuminated area and the shape of the control curve in each iteration loop. The first determining module is used to determine the first sub-individual of each body in the illuminated area based on a random step size that follows a Lévy distribution, and to determine the second sub-individual of each body in the illuminated area based on the current optimal solution of the illuminated area. The second determining module is used to determine the first sub-individual of each body in the shadow area based on the gradient information and velocity information of each body in the shadow area and the current optimal solution of the shadow area, and to determine the second sub-individual of each body in the shadow area based on the current optimal solution of the shadow area. The third determination module is used to determine each individual in the illuminated area, each individual in the shaded area, the first child of each individual in the illuminated area, the second child of each individual in the illuminated area, the first child of each individual in the shaded area, and the second child of each individual in the shaded area as all individuals in this generation, and to calculate the fitness value of all individuals in this generation. The filtering module is used to determine the complexity of the current generation's best individual based on the first indicator function, adjust the number of path points of all individuals in the next generation based on the complexity of the current generation's best individual, and filter out the better individuals of the current generation whose fitness values ​​are within a preset number before sorting as the initial individuals of the next iteration. The first indicator function is used to calculate the number of path segments on the path of the current generation's best individual that are within a preset threat area radius. The output module is used to determine the global optimal solution as the optimal obstacle avoidance path for the UAV when the number of iterations is a preset number of rounds.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A readable storage medium, the readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images