An unmanned aerial vehicle path planning method based on multi-population enhanced alpha evolution

Abstract

This invention relates to the field of unmanned aerial vehicle (UAV) control and provides a UAV path planning method based on multi-population augmented alpha evolution. The method includes: constructing a trajectory cost; determining the starting and ending coordinates of the UAV; generating an initial population based on these coordinates; selecting candidate populations from the initial population based on the trajectory cost to obtain a sampling matrix for the candidate populations; obtaining an updated population by dynamically adjusting the learning rate based on the adaptive basis vectors derived from the sampling matrix; performing particle swarm optimization based on the particle positions and movement speeds of the updated population to obtain a target population; iterating through the target population, and performing augmented migration, individual perturbation, and collaborative updates during the iteration process to obtain an iterative population; selecting an initial route from the iterative population based on the trajectory cost; and smoothing and multi-dimensional fitting of the initial route to obtain the target path. This invention can generate UAV flight paths.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a UAV path planning method based on multi-population enhanced alpha evolution. Background Technology

[0002] With the rapid development of UAV technology, multi-UAV systems have been widely applied. Compared to single-UAV missions, multi-UAV systems can significantly improve mission efficiency, expand coverage, and enhance system robustness through collaborative operations. In multi-UAV mission execution, path planning, as one of the core technologies, directly affects the mission completion efficiency and flight safety of the UAV system. Therefore, how to generate efficient, feasible, and collaborative flight paths for multiple UAVs under complex environments and multiple constraints has become one of the key research issues.

[0003] Existing research on path planning problems mainly falls into two categories: methods based on traditional optimization methods and methods based on intelligent optimization algorithms. Regarding traditional methods, search-based path planning algorithms and mathematical programming-based methods can find optimal or near-optimal solutions under certain conditions. However, these methods typically rely on the accuracy of environmental modeling and are prone to excessive computational complexity or difficulty in finding optimal solutions when facing high-dimensional state spaces, multi-constraint coupling, and dynamically changing environments, limiting their application in multi-UAV scenarios. Sampling-based path planning methods have also received widespread attention in recent years. Typical methods, such as fast random trees and their improved algorithms, construct search trees by randomly sampling in the state space, gradually approximating feasible paths. These methods do not require precise environmental modeling, possess strong flexibility and good adaptability to high-dimensional spaces, and are particularly suitable for complex or unknown environments. However, due to their strong randomness, the generated paths often suffer from unevenness or slow convergence speed, requiring further optimization and improvement in multi-constraint collaborative scenarios.

[0004] In the field of intelligent optimization methods, recent research has largely employed metaheuristic approaches such as genetic algorithms, particle swarm optimization, and ant colony optimization to solve UAV path planning problems. These methods possess strong global search capabilities, can obtain optimal solutions in complex search spaces, and have relatively low dependence on the problem model. However, existing intelligent optimization methods still have certain limitations in multi-UAV scenarios. First, regarding search capabilities, single-swarm optimization algorithms are prone to a decline in population diversity during iteration, leading to premature convergence, getting trapped in local optima, and failing to obtain the globally optimal path. Second, regarding problem complexity, compared to single-UAV path planning which mainly focuses on individual path optimization, multi-UAV path planning requires the introduction of additional collaborative constraints such as collision avoidance between UAVs, transforming the originally independent path optimization problem into a multi-agent coupled problem. Third, regarding adaptation to complex environments, when the scale of the environment expands or the number of constraints increases, the search efficiency and solution stability of existing methods decrease significantly, making it difficult to balance computational efficiency and planning quality.

[0005] Furthermore, while some improved intelligent optimization algorithms proposed in recent years, such as hybrid optimization algorithms that integrate multiple strategies, have improved path planning performance to some extent, most are still based on a single-species evolution mechanism, lacking effective information exchange and collaborative search strategies. This makes it difficult to fully utilize the complementary advantages between different search individuals, thus limiting the algorithm's performance in complex multi-modal optimization problems. Meanwhile, in multi-UAV collaborative path planning tasks, how to ensure global search capabilities while also considering local fine-grained optimization capabilities remains a critical issue that urgently needs to be addressed. Summary of the Invention

[0006] This invention aims to at least solve one of the technical problems existing in related technologies. To this end, this invention provides a drone path planning method based on multi-population enhanced alpha evolution, which optimizes the flight paths of multiple drones while ensuring safety.

[0007] This invention provides a UAV path planning method based on multi-population enhanced alpha evolution, comprising: S1: Determine the UAV operating environment, establish the UAV operating cost function based on the UAV operating environment, and construct the trajectory cost through the UAV operating cost function; S2: Determine the starting coordinates and ending coordinates of the drone, and generate an initial population based on the starting coordinates and ending coordinates of the drone; S3: Select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated populations to obtain the target population. S4: Iterate through the target population, and during the iteration process, perform enhanced migration, individual perturbation and collaborative update on the target population to obtain the iterative population. Select the initial route in the iterative population based on the trajectory cost. S5: The initial route is smoothed and fitted in multiple dimensions to obtain the target path, which the UAV uses as its flight path.

[0008] According to the UAV path planning method based on multi-population enhanced alpha evolution provided by the present invention, step S1 further includes: S11: Determine the operating environment of the UAV, perform grid-based modeling of the UAV operating environment, and obtain a grid map; S12: Based on the grid map, establish an obstacle cost function, construct a kinematic cost function and a multi-machine cooperative cost function. The UAV operation cost function includes the obstacle cost function, the kinematic cost function and the multi-machine cooperative cost function. Construct the trajectory cost through the UAV operation cost function.

[0009] According to the present invention, a method for UAV path planning based on multi-population enhanced alpha evolution is provided. In step S2, the starting coordinates and ending coordinates of multiple UAVs are determined, a search space is constructed according to the UAV operating environment, a random uniform distribution is adopted in the search space, and an initial population is generated according to the starting coordinates and ending coordinates of the UAVs.

[0010] According to the UAV path planning method based on multi-population enhanced alpha evolution provided by the present invention, step S3 further includes: S31: Calculate the fitness of the initial population using the trajectory cost, and select the candidate population from the initial population based on the fitness; S32: Construct the sampling matrix using the candidate population, extract the diagonal elements of the sampling matrix, generate an adaptive basis vector based on the diagonal elements, and obtain the updated population by dynamically adjusting the learning rate using the adaptive basis vector; Obtain the particle positions and movement speeds of the updated population, update the movement speed of the updated population based on the particle positions and movement speeds to obtain the updated particle speeds, and optimize the particle swarm based on the updated particle speeds to obtain the target population.

[0011] According to the present invention, a UAV path planning method based on multi-population enhanced alpha evolution is provided. In step S4, the target population is used to replace the initial population and step S3 is repeated to iterate the target population. During the iteration process, an elite population is selected from the target population and added to other target populations to complete the enhanced migration.

[0012] According to the UAV path planning method based on multi-population enhanced alpha evolution provided by the present invention, in step S5, path points are obtained through the initial route, and the path points are smoothed in multiple dimensions using the cubic smoothing spline method to obtain the target path.

[0013] This invention also provides a UAV path planning system based on multi-population enhanced alpha evolution, comprising: The trajectory cost function module is used to determine the UAV's operating environment, establish the UAV operating cost function based on the UAV's operating environment, and construct the trajectory cost through the UAV operating cost function. Initial Population Module: Used to determine the starting coordinates and ending coordinates of the UAV, and generate the initial population based on the starting coordinates and ending coordinates of the UAV; The target population module is used to select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated population to obtain the target population. Initial route module: used to iterate over the target population, and to perform enhanced migration, individual perturbation and collaborative update of the target population during the iteration process to obtain the iterative population. The initial route is selected in the iterative population based on the trajectory cost. Target path module: used to smooth the initial route and fit it in multiple dimensions to obtain the target path, which the UAV will use as its flight path.

[0014] The present invention also 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 a UAV path planning method based on multi-population augmented alpha evolution as described above.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a UAV path planning method based on multi-population augmented alpha evolution as described above.

[0016] The present invention also provides a computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions that, when executed by a computer, enable the computer to perform the steps of any of the above-described methods for unmanned aerial vehicle path planning based on multi-population enhanced alpha evolution.

[0017] The above-described one or more technical solutions in the embodiments of the present invention have at least one of the following technical effects: This invention provides a UAV path planning method based on multi-population enhanced alpha evolution. It evaluates and constrains the UAV's flight path through trajectory cost, and optimizes the population through adaptive basis vectors and particle swarm optimization to obtain a better target population. During the iteration process of the target population, the efficiency of the iteration is improved by enhancing migration, individual perturbation and collaborative update, and finally the optimal target path is obtained smoothly.

[0018] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a UAV path planning method based on multi-population enhanced alpha evolution provided by the present invention.

[0021] Figure 2 This is a schematic diagram of the structure of a UAV path planning system based on multi-population enhanced alpha evolution provided by the present invention.

[0022] Figure 3 This is a schematic diagram of the structure of a drone path planning device based on multi-population enhanced alpha evolution provided by the present invention.

[0023] Figure label: 100. Trajectory Cost Function Module; 200. Initial Population Module; 300. Target Population Module; 400. Initial Route Module; 500. Target Path Module; 810. Processor; 820. Communication Interface; 830. Memory; 840. Communication Bus. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The following embodiments are used to illustrate this invention but cannot be used to limit the scope of this invention.

[0025] In the description of the embodiments of the present invention, it should be noted that the terms "first", "second" and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0026] In the description of the embodiments of the present invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in the embodiments of the present invention based on the specific circumstances.

[0027] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0028] The following is combined Figures 1 to 3 Specific embodiments of the present invention are described below. Figure 1 A flowchart illustrating a UAV path planning method based on multi-population enhanced alpha evolution provided by this invention includes: S1: Determine the UAV operating environment, establish the UAV operating cost function based on the UAV operating environment, and construct the trajectory cost through the UAV operating cost function; Furthermore, the objective of this stage is to establish the UAV operation cost function, thereby constructing the trajectory cost. Specifically, step S1 further includes: S11: Determine the operating environment of the UAV, perform grid-based modeling of the UAV operating environment, and obtain a grid map; S12: Based on the grid map, establish an obstacle cost function, construct a kinematic cost function and a multi-machine cooperative cost function. The UAV operation cost function includes the obstacle cost function, the kinematic cost function and the multi-machine cooperative cost function. Construct the trajectory cost through the UAV operation cost function.

[0029] The specific implementation method for the above steps in this embodiment is as follows: First, the environment in which the UAV will fly needs to be determined, i.e., the UAV operating environment. Then, a 3D grid map is used to discretize and model the UAV operating environment, resulting in a grid map. Specifically, first, the scale of the 3D grid map is determined based on the actual dimensions of the UAV operating environment, and the entire space is uniformly divided into several cubic grid cells. The side length of each grid cell can be set according to the planning accuracy requirements. To simulate typical obstacles in complex environments, obstacles are set as cylinders with defined diameters and heights in the grid map.

[0030] Next, an obstacle cost function needs to be established based on the grid map. Specifically: Where R is the threat radius of the obstacle, which is the radius of the obstacle plus a threat value determined empirically; r is the radius of the obstacle; and d is the distance from the i-th path point of the UAV to the center coordinates of the obstacle. The obstacle threat score is given for the i-th path point, and N is the total number of path points. Score the cost of obstacles.

[0031] In addition, kinematic cost functions need to be constructed, including path length cost function, pitch angle cost function, yaw angle cost function, and flight altitude cost function. The path length cost function is as follows: in, Let i be the coordinates of the (i+1)th path point. Let i be the coordinates of the i-th path point. The path cost score is used to limit the total length of the path during the drone's mission execution, ensuring that the path has good energy efficiency and execution efficiency while meeting mission requirements. An excessively long path will cause additional increase in mission completion time and energy consumption, and may also cause problems such as drone battery depletion and forced mission interruption.

[0032] The pitch angle cost function is: in, Let be the pitch angle of the i-th path point. This represents the height increment of the i-th path point. The x-coordinate increment of the i-th path point. This represents the increment of the ordinate of the i-th path point. Let be the pitch angle of the (i-1)th path point. The pitch angle cost score is calculated using arctan(), which calculates the arctangent value within the parentheses. The pitch angle cost function is defined here because the vertical angle variation of the UAV during flight between adjacent path nodes must be within a specified range. Excessive pitch angles increase flight losses and may even lead to loss of control and crashes. Therefore, in path planning, it is essential to strictly control excessive pitch angle variations to ensure flight safety and stability, and reduce mechanical damage and energy consumption caused by excessive pitch angles.

[0033] The yaw cost function is: Here, arccos() calculates the arccosine value within the parentheses. Let i be the horizontal projection direction vector of the i-th path point. Let be the horizontal projection direction vector of the (i-1)th path point. Let yaw angle be the yaw angle of the i-th path point. Scoring is given for yaw angle cost. Here, when the drone flies between adjacent waypoints, the turning angle must be kept within a specified range. The smaller the turning radius of the drone, the larger the required yaw angle, making it easier to lose balance during the turn and increasing flight risk; conversely, a larger turning radius allows the drone to fly more smoothly and reduces the possibility of loss of balance.

[0034] The flight altitude cost function is: in, Let be the height of the i-th path point. The preset maximum flight altitude, The preset minimum flight altitude, The score is the flight altitude of the i-th path point. The altitude cost score is used to ensure that the drone flies at a safe altitude and avoids collisions with the ground or other flying objects.

[0035] Furthermore, in multi-drone collaborative operation scenarios, in order to prevent multiple drones from overlapping paths or flying close together during flight by planning reasonable spatial intervals, thereby effectively avoiding aircraft collisions, it is necessary to construct a multi-drone collaborative cost function: in, Let be the distance between the i-th path point at time t and the j-th path point of another adjacent UAV. For a safe distance, The multi-aircraft cooperative cost score is then calculated. Finally, the path cost score, pitch angle cost score, yaw angle cost score, flight altitude cost score, obstacle cost score, and multi-aircraft cooperative cost score are weighted and summed to obtain the trajectory cost of the path point.

[0036] S2: Determine the starting coordinates and ending coordinates of the drone, and generate an initial population based on the starting coordinates and ending coordinates of the drone; Furthermore, the objective of this stage is to generate an initial population. Specifically, in step S2, the starting coordinates and ending coordinates of multiple UAVs are determined, a search space is constructed based on the UAV operating environment, a random and uniform distribution is adopted in the search space, and an initial population is generated based on the starting coordinates and ending coordinates of the UAVs.

[0037] The specific implementation method for the above steps in this embodiment is as follows: First, the starting and ending coordinates of each drone need to be determined. Then, the drone's operating environment is used as the search space for its coordinates. Within this search space, a set of initial path points needs to be generated for each drone, thus forming multiple independent initial populations for each drone. Next, upper and lower bounds are determined for each path point's parameters, ensuring that these parameters are randomly and uniformly distributed within these bounds. This process generates several feasible candidate paths for each drone, thus obtaining the initial population.

[0038] S3: Select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated populations to obtain the target population. Furthermore, the objective of this stage is to obtain candidate populations, thereby dynamically adjusting the learning rate to obtain updated populations, and finally performing particle swarm optimization to obtain the target population. Specifically, step S3 further includes: S31: Calculate the fitness of the initial population using the trajectory cost, and select the candidate population from the initial population based on the fitness; S32: Construct the sampling matrix using the candidate population, extract the diagonal elements of the sampling matrix, generate an adaptive basis vector based on the diagonal elements, and obtain the updated population by dynamically adjusting the learning rate using the adaptive basis vector; Obtain the particle positions and movement speeds of the updated population, update the movement speed of the updated population based on the particle positions and movement speeds to obtain the updated particle speeds, and optimize the particle swarm based on the updated particle speeds to obtain the target population.

[0039] The specific implementation method for the above steps in this embodiment is as follows: First, the fitness of candidate paths in each initial population is calculated using the trajectory cost calculation method described above. Here, a higher trajectory cost corresponds to a lower fitness. Then, a batch of candidate paths with higher fitness is selected from the initial population as the candidate population. Next, a sampling matrix is ​​constructed based on the parameters of each candidate path in the candidate population. The diagonal elements of the sampling matrix are extracted, and an adaptive basis vector X is generated using these elements. Subsequently, when updating the population based on the adaptive basis vector, the learning rate needs to be dynamically adjusted. This ensures greater exploration capability in the early stages of iteration and gradually enhances convergence in later stages, thereby obtaining the control parameters. The control parameters can control the direction and magnitude of population generation and updates. Furthermore, a random step size for the k-th update path, obtained from a pre-constructed perturbation matrix, needs to be introduced. This allows us to obtain the kth update path. : Where α is the coefficient of the adaptive basis vector. This is the path before the update of the k-th update path selected from the initial population. The initial path selected from the initial population has a fitness level superior to the path before the update. The initial path is selected from the initial population whose fitness is worse than the path before the update. If the fitness of the updated path is better than the path before the update, it is used to replace the path before the update; otherwise, the path before the update remains unchanged. After the path is updated, boundary correction is performed by halving the distance for any out-of-bounds variables to ensure the feasibility of the solution. This process of replacing the initial population several times yields the updated population.

[0040] During the process of obtaining the updated population, the position of the particle on the k-th update path in the updated population is obtained. and movement speed This updates the movement speed of the population and obtains the updated particle velocity. : Where w is the inertia weight, For individual learning factors, As a global learning factor, For individual coefficients, For global coefficients, As a preset global optimum, The optimal value for each individual is predetermined. The movement speed of the updated path is adjusted based on the updated movement speed, allowing the updated path to perform a local search near the current optimal solution. This fine-tuning of the updated path gradually converges it to a local optimum, improving its accuracy and stability, eliminating local suboptimal structures within the updated path, and thus adjusting the movement speed of the updated path based on the updated particle velocity. This completes particle swarm optimization and yields the target population. Here, the inertia weight decreases over time to facilitate the search from exploration to convergence.

[0041] S4: Iterate through the target population, and during the iteration process, perform enhanced migration, individual perturbation and collaborative update on the target population to obtain the iterative population. Select the initial route in the iterative population based on the trajectory cost. Furthermore, the objective of this stage is to enhance migration, individual perturbation, and collaborative updates of the target population to obtain an iterative population, thereby obtaining the initial route. Specifically, in step S4, the target population is used to replace the initial population, and step S3 is repeated to iterate the target population. During the iteration process, an elite population is selected from the target population and added to other target populations to complete the enhanced migration.

[0042] The specific implementation method for the above steps in this embodiment is as follows: After obtaining multiple target populations, the initial populations are replaced with these target populations, and step S3 is repeated iteratively. During the iteration process, every 50 iterations, a high-fitness update path is selected from the obtained target populations as an elite population. This elite population is then transferred to other target populations, replacing the lower-fitness populations. This completes the augmentation transfer, and the iteration continues. During iteration, the parameters that each target population focuses on optimizing can differ. Furthermore, individual perturbations and collaborative updates are also required during the iteration process.

[0043] Individual perturbation involves randomly adjusting some parameters of the paths in the update population using small Gaussian perturbations, introducing some variation while maintaining the original favorable structure, thereby enhancing population diversity. Collaborative update involves sharing high-quality parameters from the update population with high overall fitness. By integrating high-quality information from other populations, each population can share global search experience while maintaining independent search capabilities, further adjusting the iterative process of the target population. Finally, the iteration ends, resulting in an iterative population. This improves the global search capability, convergence speed, and solution diversity of the iterative population, effectively avoiding population homogenization and thus enhancing the collaborative optimization capability of multiple populations and the overall performance of the method. Finally, the update path with the highest fitness is selected from the iterative population as the initial route.

[0044] S5: The initial route is smoothed and fitted in multiple dimensions to obtain the target path, which the UAV uses as its flight path.

[0045] Furthermore, the objective of this stage is to obtain the target path so that the UAV can use it as its flight path. Specifically, in step S5, path points are obtained from the initial route, and these path points are smoothed in multiple dimensions using a cubic smoothing spline method to obtain the target path.

[0046] The specific implementation method for the above steps in this embodiment is as follows: Here, a series of path points can be extracted from the initial route. These path points need to be connected, and the connections are smoothed in various dimensions using a cubic smoothing spline method. This reduces local discontinuities and abrupt changes in direction while ensuring the overall path structure remains largely unchanged. During the smoothing process, the balance between path fidelity and smoothness is controlled by adjusting the smoothing factor. This ensures that the path maintains the overall trend of the original planning result while effectively eliminating local oscillations introduced by discrete optimization, thus obtaining the target path. The UAV can then use this target path as its flight route.

[0047] The effectiveness of the proposed UAV path planning method based on multi-population enhanced alpha evolution is also verified here. Multiple methods are compared with this method in two different constructed scenarios, namely scenario 1 and scenario 2. The comparison results are shown in Table 1: Table 1. Comparison of target path generation performance between this method and other methods.

[0048] Among them, GWO, PSO, AE, DOA, CPO, MSO, and ESC are all existing path planning methods. The average value is the average trajectory cost of the target path, the standard deviation is the standard deviation of the trajectory cost of the target path, and the average ranking is the ranking of each method based on the average value and standard deviation. It can be seen that, under different drone sorties and scenarios, the method provided by this invention can achieve good results and the highest ranking in various scenarios and sorties.

[0049] The following describes a UAV path planning device based on multi-population enhanced alpha evolution provided by the present invention. The UAV path planning device based on multi-population enhanced alpha evolution described below and the UAV path planning method based on multi-population enhanced alpha evolution described above can be referred to in correspondence.

[0050] Figure 2An example is a schematic diagram of the structure of a UAV path planning system based on multi-population augmented alpha evolution, such as... Figure 2 As shown, a method for executing a UAV path planning method based on multi-population augmented alpha evolution, as described above, includes: Trajectory Cost Function Module 100: Used to determine the UAV operating environment, establish the UAV operating cost function based on the UAV operating environment, and construct the trajectory cost through the UAV operating cost function; Initial Population Module 200: Used to determine the starting coordinates and ending coordinates of the UAV, and generate an initial population based on the starting coordinates and ending coordinates of the UAV; Target population module 300: It is used to select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated populations to obtain the target population. Initial route module 400: It is used to iterate the target population, and during the iteration process, enhance migration, individual perturbation and collaborative update of the target population to obtain the iterative population. The initial route is selected in the iterative population based on the trajectory cost. Target path module 500: Used to smooth the initial route and fit it in multiple dimensions to obtain the target path, which the UAV will use as its flight path.

[0051] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call a computer program in the memory 830 to execute a UAV path planning method based on multi-population enhanced alpha evolution, the method including: S1: Determine the UAV operating environment, establish the UAV operating cost function based on the UAV operating environment, and construct the trajectory cost through the UAV operating cost function; S2: Determine the starting coordinates and ending coordinates of the drone, and generate an initial population based on the starting coordinates and ending coordinates of the drone; S3: Select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated populations to obtain the target population. S4: Iterate through the target population, and during the iteration process, perform enhanced migration, individual perturbation and collaborative update on the target population to obtain the iterative population. Select the initial route in the iterative population based on the trajectory cost. S5: The initial route is smoothed and fitted in multiple dimensions to obtain the target path, which the UAV uses as its flight path.

[0052] Furthermore, when the computer program in the aforementioned memory 830 can be implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0053] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein when the program instructions are executed by a computer, the computer is able to execute a UAV path planning method based on multi-population enhanced alpha evolution provided by the above methods, the method comprising: S1: Determine the UAV operating environment, establish the UAV operating cost function based on the UAV operating environment, and construct the trajectory cost through the UAV operating cost function; S2: Determine the starting coordinates and ending coordinates of the drone, and generate an initial population based on the starting coordinates and ending coordinates of the drone; S3: Select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated populations to obtain the target population. S4: Iterate through the target population, and during the iteration process, perform enhanced migration, individual perturbation and collaborative update on the target population to obtain the iterative population. Select the initial route in the iterative population based on the trajectory cost. S5: The initial route is smoothed and fitted in multiple dimensions to obtain the target path, which the UAV uses as its flight path.

[0054] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the aforementioned method for unmanned aerial vehicle path planning based on multi-population enhanced alpha evolution, the method comprising: S1: Determine the UAV operating environment, establish the UAV operating cost function based on the UAV operating environment, and construct the trajectory cost through the UAV operating cost function; S2: Determine the starting coordinates and ending coordinates of the drone, and generate an initial population based on the starting coordinates and ending coordinates of the drone; S3: Select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated populations to obtain the target population. S4: Iterate through the target population, and during the iteration process, perform enhanced migration, individual perturbation and collaborative update on the target population to obtain the iterative population. Select the initial route in the iterative population based on the trajectory cost. S5: The initial route is smoothed and fitted in multiple dimensions to obtain the target path, which the UAV uses as its flight path.

[0055] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0056] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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; and these 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 the present invention.

Claims

1. A method for unmanned aerial vehicle path planning based on multi-population enhanced alpha evolution, characterized in that, include: S1: Determine the UAV operating environment, establish the UAV operating cost function based on the UAV operating environment, and construct the trajectory cost through the UAV operating cost function; S2: Determine the starting coordinates and ending coordinates of the drone, and generate an initial population based on the starting coordinates and ending coordinates of the drone; S3: Select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated populations to obtain the target population. S4: Iterate through the target population, and during the iteration process, perform enhanced migration, individual perturbation and collaborative update on the target population to obtain the iterative population. Select the initial route in the iterative population based on the trajectory cost. S5: The initial route is smoothed and fitted in multiple dimensions to obtain the target path, which the UAV uses as its flight path.

2. The method of claim 1, wherein, Step S1 further includes: S11: Determine the operating environment of the UAV, perform grid-based modeling of the UAV operating environment, and obtain a grid map; S12: Based on the grid map, establish an obstacle cost function, construct a kinematic cost function and a multi-machine cooperative cost function. The UAV operation cost function includes the obstacle cost function, the kinematic cost function and the multi-machine cooperative cost function. Construct the trajectory cost through the UAV operation cost function.

3. The method of claim 1, wherein, In step S2, the starting coordinates and ending coordinates of multiple drones are determined, a search space is constructed based on the drone operating environment, a random uniform distribution is adopted in the search space, and an initial population is generated based on the starting coordinates and ending coordinates of the drones.

4. The method of claim 1, wherein, Step S3 further includes: S31: Calculate the fitness of the initial population using the trajectory cost, and select the candidate population from the initial population based on the fitness; S32: Construct the sampling matrix using the candidate population, extract the diagonal elements of the sampling matrix, generate an adaptive basis vector based on the diagonal elements, and obtain the updated population by dynamically adjusting the learning rate using the adaptive basis vector; Obtain the particle positions and movement speeds of the updated population, update the movement speed of the updated population based on the particle positions and movement speeds to obtain the updated particle speeds, and optimize the particle swarm based on the updated particle speeds to obtain the target population.

5. The method of claim 1, wherein, In step S4, the target population is used to replace the initial population, and step S3 is repeated to iterate the target population. During the iteration process, an elite population is selected from the target population and added to other target populations to complete the enhanced migration.

6. The method of claim 1, wherein, In step S5, path points are obtained through the initial route, and the path points are smoothed in multiple dimensions using the cubic smoothing spline method to obtain the target path.

7. A multi-population enhanced alpha evolution based UAV path planning system for performing a multi-population enhanced alpha evolution based UAV path planning method according to any one of claims 1 to 6, characterized in that, include: The trajectory cost function module is used to determine the UAV's operating environment, establish the UAV operating cost function based on the UAV's operating environment, and construct the trajectory cost through the UAV operating cost function. Initial Population Module: Used to determine the starting coordinates and ending coordinates of the UAV, and generate the initial population based on the starting coordinates and ending coordinates of the UAV; The target population module is used to select candidate populations from the initial population based on trajectory cost, obtain the sampling matrix of the candidate populations, obtain adaptive basis vectors based on the sampling matrix and dynamically adjust the learning rate to obtain the updated population, and perform particle swarm optimization based on the particle positions and movement speeds of the updated population to obtain the target population. Initial route module: used to iterate over the target population, and to perform enhanced migration, individual perturbation and collaborative update of the target population during the iteration process to obtain the iterative population. The initial route is selected in the iterative population based on the trajectory cost. Target path module: used to smooth the initial route and fit it in multiple dimensions to obtain the target path, which the UAV will use as its flight path.

8. 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 unmanned aerial vehicle path planning method based on multi-population enhanced alpha evolution as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the unmanned aerial vehicle path planning method based on multi-population enhanced alpha evolution as described in any one of claims 1 to 6.

10. A computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, characterized in that, When the program instructions are executed by the computer, the computer is able to perform the steps of a drone path planning method based on multi-population enhanced alpha evolution as described in any one of claims 1 to 6.

Images