UAV Autonomous Path Planning Method Based on Visual Perception Module

Abstract

An autonomous path planning method for drones based on a visual perception module, which relates to the technical field of drone inspection. The method includes the following steps: Step 1: Collect the coordinates and image features of inspection targets and obstacles in the inspection area, and build an environmental model of the drone inspection area; Step 2: Based on the environmental model, evaluate the collision risk level of the obstacle area according to the maneuverability characteristics of the drone; Step 3: Based on the risk level, establish a fitness function for the safe planned path of the drone; Step 4: Based on the improved particle swarm algorithm, optimize the fitness function to obtain the optimal planned path for the current inspection area. The present invention improves the safety and reliability of drones when performing inspection tasks.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV inspection, and particularly to a method for autonomous path planning of UAVs based on a visual perception module. Background Art

[0002] At the forefront of modern technology, unmanned aerial vehicles (UAVs) have become key tools for cross-industry applications, from agricultural monitoring to emergency rescue, and to urban planning and environmental science. Their versatility and efficiency have been widely recognized. However, the autonomous navigation of UAVs faces severe challenges in complex environments, especially in signal-constrained areas such as urban canyons and dense forests. Traditional GPS and preset path planning methods often fail to ensure real-time, accurate positioning and dynamic obstacle avoidance.

[0003] In recent years, the progress of visual perception technology has opened up new ways for the autonomous flight of UAVs. By carrying high-resolution cameras and advanced image processing algorithms, UAVs can capture and analyze environmental data in real time, including terrain features, obstacle positions, and moving target trajectories. This ability has greatly improved the adaptability of UAVs in unknown or rapidly changing environments, enabling them to instantaneously plan safe and effective flight routes and thus maintain stable operation under various conditions. However, to achieve true autonomous flight, a series of technical challenges still need to be overcome, including but not limited to the real-time processing and analysis of visual data, precise positioning and navigation in complex environments, real-time obstacle avoidance for dynamic obstacles, and efficient communication with cloud or ground stations.

[0004] In the prior art, for example, in the patent application document with the application number 202410185850.7 and the title "An UAV path planning algorithm for detecting insulators on power towers", path planning is carried out for insulators on the tower, and the scenario is single; moreover, the particle swarm algorithm used for solving is a basic particle swarm algorithm, and the algorithm has limitations such as slow convergence speed and easy to fall into local optimum.

[0005] For example, in the patent application document with the application number 201811583287.X and the title "A global multi-objective particle swarm planning method for UAV three-dimensional flight tracks", the grid method is used to discretize map information for subsequent calculations, but the grid method brings the problem of high computational complexity, requires too high computational power for the real-time obstacle avoidance of UAVs, lacks research on the impact of UAV flight states on path planning accuracy, and lacks research on path planning in the case of dynamic obstacles, with low reliability. Summary of the Invention

[0006] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0007] The technical solution of the present invention is: an autonomous path planning method for drones based on a visual perception module, comprising the following steps:

[0008] Step 1: Collect the coordinates and image features of inspection targets and obstacles in the inspection area, and build an environmental model of the drone inspection area;

[0009] Step 2: Based on the environmental model, evaluate the collision risk level of the obstacle area according to the maneuvering characteristics of the drone;

[0010] Step 3: Based on the risk level, establish a fitness function for the safe planned path of the drone;

[0011] Step 4: Based on the improved particle swarm algorithm, optimize the fitness function to obtain the optimal planned path in the current inspection area.

[0012] In Step 1, use the drone to carry a high-resolution binocular camera to scan the inspection area comprehensively to obtain high-definition image and video data; at the same time, use the image processing technology of edge detection to extract the feature information of inspection targets and obstacles from the images, including shape and size;

[0013] Use GPS positioning technology to record the flight position coordinates of the drone, the coordinates of inspection targets and the geographical location coordinates of obstacles;

[0014] Utilize the collected image and coordinate data, and adopt positioning and mapping technology to combine the coordinates with the image features to form an environmental model.

[0015] In Step 1, it also includes: after normalizing and cross-validating the collected image and coordinate data, establish a data indexing and retrieval mechanism, and store the relevant data in the inspection database.

[0016] In Step 2, according to the maneuvering characteristics of the drone, evaluate the maneuvering ability of the drone, and then combine with the environmental model to evaluate the collision risk level of the obstacle area.

[0017] Collect the basic performance parameters of the inspection drone, including the maximum flight speed, maximum acceleration, minimum turning radius, maximum climb rate and descent rate,

[0018] Among them,

[0019]

[0020] In the formula, is the flight speed of the drone at time t, v max is the maximum speed of the drone flight, the flight speed of the drone at time t + Δt, is the acceleration of the drone in the Δt period;

[0021] r t UAV is the turning radius of the UAV during the t period, r min is the minimum turning radius of the UAV are the climb rate and descent rate of the UAV during the Δt period respectively is the flight altitude of the UAV at the t + Δt period is the flight altitude of the UAV at the t period

[0022] Analyze the maneuvering range of the UAV in different flight modes such as hovering, straight flight, and turning, and evaluate the maneuverability according to the maneuvering range:

[0023]

[0024] Among them, 1, 2, and 3 are the levels of maneuverability

[0025] Model the environmental characteristics in the inspection area as similar mathematical models according to the edge characteristics in sequence, and calculate the minimum safety distance d between the UAV and any point in the three-dimensional space t , expressed as:

[0026]

[0027] Among them is the coordinate of the center point of the UAV; (x, y, z) is any point on the plane; R is the radius size when the UAV is mathematically modeled as a circle

[0028] Combine the speed and turning radius of the UAV at the t period, as well as the minimum safety distance from the obstacle, to evaluate the risk level of the flight area

[0029] Among them

[0030]

[0031] Among them, Level is the risk level, L low 、L mid 、L high are low risk, medium risk, and high risk respectively, D min is the maximum centripetal acceleration, d min is the minimum safety distance between the UAV and the obstacle

[0032] The fitness function is:

[0033]

[0034] Among them, F fitness is the fitness function, and P is the flight power of the UAV

[0035] In step 4, the constructed environmental model is used to simulate the inspection area for the UAV path planning. In the environmental model, random dynamic obstacles are added. The vision perception module carried by the UAV is used to detect new obstacles in the environment, and range measurement, edge detection and model update are carried out online.

[0036] Based on the particle swarm optimization algorithm and the maneuverability of the UAV, it helps the UAV avoid obstacles, select a suitable flight route, and evaluate the quality of the path planned by the algorithm according to the value of the fitness function and the time of algorithm iteration, so as to optimize the flight strategy.

[0037] According to the complexity of the inspection area, the initial population size and particle positions are set. Each particle represents a possible flight path.

[0038] Based on the local optimal solution and the global optimal solution, the search direction of the particle swarm is guided.

[0039] According to the number of iterations and the current optimization situation, the inertia weight and cognitive factor are adaptively adjusted to balance the global search and local search capabilities, and the speed and position of each particle are updated.

[0040] Step 4 includes:

[0041] S41: Initialize the population size, maximum number of iterations, inertia weight and cognitive factor, as well as particle speed and position;

[0042] S42: The particle calculates the fitness value according to the constructed fitness function.

[0043] S43: Update the inertia weight and cognitive factor according to the improved inertia weight and cognitive factor formulas; at the same time, introduce a perturbation factor; update the speed and position of the particle according to the speed and the improved position formula.

[0044] S44: Update the individual optimal and global optimal: Compare the current fitness value of the particle with the individual historical optimal value. If the current value is better, update the individual optimal position to the current position; among the individual optimal positions of all particles, determine the global optimal position, and find the individual optimal position of the particle with the best fitness value as the global optimal position.

[0045] S45: Judge the termination condition. If the maximum number of iterations is reached, the algorithm terminates and outputs the global optimal position and fitness value; otherwise, return to S42 to continue the iteration.

[0046] The present invention has the following beneficial effects:

[0047] 1) The environmental modeling is achieved through visual perception and GPS positioning, which not only improves the high-fidelity simulation of the inspection area, but also enhances the perception of environmental changes during the flight of the UAV, that is, the recognition and positioning of dynamic obstacles, thereby dynamically improving the flight safety of the UAV in complex environments;

[0048] 2) According to the minimum safe distance between the UAV and the obstacles, and integrating the maneuverability of the UAV, the risk level of the UAV inspection area is evaluated, and safety warnings and the UAV maneuverability coefficient are provided in path planning;

[0049] 3) A fitness function for flight safety and energy consumption efficiency is constructed, constrained by the maneuvering characteristics of the UAV and the risk level of the flight area, to better cope with environmental changes. Based on the improved particle swarm optimization algorithm, the path planning is solved to achieve the robustness and stability of the planned path.

[0050] The present invention improves the safety and reliability of the UAV when performing inspection tasks. Description of the Drawings

[0051] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. In the drawings, the parts are not necessarily drawn to actual scale.

[0052] Figure 1 It is the flowchart of the method of the present invention. Detailed Embodiments

[0053] The present invention is as Figure 1 shown, and the UAV autonomous path planning method based on the visual perception module includes the following steps:

[0054] Step 1: Collect the coordinates and image features of the inspection targets and obstacles in the inspection area, and build an environmental model of the UAV inspection area;

[0055] Step 2: Based on the environmental model, evaluate the collision risk level of the obstacle area according to the maneuvering characteristics of the UAV;

[0056] Step 3: Based on the risk level, establish a fitness function for the safe planned path of the UAV;

[0057] Step 4: Based on the improved particle swarm optimization algorithm, optimize the fitness function to obtain the optimal planned path for the current inspection area.

[0058] The present invention realizes environmental modeling through visual perception and GPS positioning, which not only improves the high-fidelity simulation of the inspection area, but also enhances the perception of environmental changes during the flight of the UAV, that is, the recognition and positioning of dynamic obstacles, thereby dynamically improving the flight safety of the UAV in complex environments;

[0059] According to the minimum safe distance between the UAV and the obstacle, and comprehensively considering the maneuverability of the UAV, evaluate the risk level of the UAV inspection area, and provide safety warnings and UAV maneuverability coefficient in path planning;

[0060] Construct a fitness function for flight safety and energy consumption efficiency, with the maneuverability characteristics of the UAV and the risk level of the flight area as constraints, better cope with environmental changes, solve path planning based on an improved particle swarm algorithm, and realize the robustness and stability of the planned path.

[0061] The specific steps are as follows:

[0062] Step 1: Collect and process the coordinates and images of inspection targets and obstacles in the inspection area, build an environmental model of the inspection area, and provide accurate environmental information for path planning:

[0063] Use the high-resolution binocular camera carried by the UAV to conduct an all-round scan of the inspection area to obtain high-definition image and video data; at the same time, use image processing technology of edge detection to extract feature information such as the shape and size of inspection targets and obstacles from the images;

[0064] Use GPS positioning technology to record the flight position coordinates of the UAV, the coordinates of inspection targets, and the geographical location coordinates of obstacles.

[0065] In order to quickly access and query specific inspection target or obstacle information, after processing the above data such as normalization and cross-validation, establish a data indexing and retrieval mechanism, and store the relevant data in the inspection database for subsequent update and analysis.

[0066] Utilize the collected image and coordinate data, and adopt Simultaneous Localization and Mapping (SLAM) technology to combine coordinates with image features to build a complete environmental model including terrain, inspection targets, obstacles, etc.

[0067] Specifically, in Step 2, comprehensively consider the maneuverability characteristics of the UAV and environmental information to evaluate the collision risk of the obstacle area:

[0068] Collect the basic performance parameters (i.e., general basic parameters) of the UAV used for inspection, including maximum flight speed, maximum acceleration, minimum turning radius, maximum climb rate and descent rate, etc.:

[0069]

[0070] Among them, is the flight speed of the UAV at time t, v max is the maximum speed of the UAV flight, is the acceleration of the drone during the time period Δt, r t UAV is the turning radius of the drone during the time period t, r min is the minimum turning radius of the drone are the climb rate and descent rate of the drone during the time period Δt, respectively is the flight altitude of the drone at the time t + Δt;

[0071] Analyze the maneuvering range of the drone in different flight modes such as hovering, straight flight, and turning, and evaluate the maneuverability according to the maneuvering range:

[0072]

[0073] Define the maneuverability of the drone according to whether the drone is in a flying and turning state. Among them, 1, 2, and 3 are the levels of maneuverability, with level 1 being the highest and level 3 being the lowest. When performing subsequent calculations for different maneuvering levels, they can be quantified as values in different intervals.

[0074] Evaluating the maneuverability of the drone is mainly to evaluate the risk degree of the drone's safe avoidance of obstacles. When the drone adjusts its path in the face of obstacles, it needs to switch flight parameters. In the three conventional states of flying, turning, and hovering, the time costs required to adjust flight parameters are inconsistent.

[0075] Considering that the time required for flight parameter switching is not instantaneous, it is necessary to evaluate the risk brought by the adjustment time during obstacle avoidance.

[0076] Combining the mathematical model of the obstacle and the maneuverability of the drone, evaluate the risk level of the drone's obstacle avoidance in this state (when the maneuvering level is low, it is necessary to increase the maneuverability coefficient of the drone to increase the shortest distance between the drone and the obstacle to plan a safe path for the drone's flight).

[0077] The inspection environment mainly includes inspection targets and obstacles (i.e., non-inspection targets), and mathematically morphological models are established for environmental features according to edge features.

[0078] Common obstacle forms, such as: buildings are modeled as cuboids, trees are modeled as cones, transmission poles are modeled as cylinders, etc.; the actual objects collected are compared with mathematical models through mathematical modeling to facilitate subsequent path planning.

[0079] Model the environmental features in the inspection area as similar mathematical models in sequence according to edge features, and calculate the minimum distance between the drone and the obstacle based on this. Judge the collision probability according to the maneuverability of the drone, and divide it into low-risk, medium-risk, and high-risk areas accordingly:

[0080] According to the morphological characteristics of the UAV, such as the minimum safety distance d between a general quadrotor UAV and any point in three-dimensional space t It is expressed as:

[0081]

[0082] wherein, is the coordinate of the center point of the UAV; (x, y, z) is any point on the plane. When this point falls on the obstacle, it simultaneously satisfies the mathematical model of the obstacle itself; R is the radius size when the UAV is mathematically modeled as a circle;

[0083] Combined with the speed, turning radius of the UAV in the t period, and the minimum safety distance from the obstacle, the risk level of the flight area is evaluated.

[0084] wherein,

[0085]

[0086] wherein, Level is the risk level (i.e., the UAV maneuverability coefficient), d t (maneuverability) is the mathematical relationship between the minimum safety distance (between the UAV and the obstacle) and the UAV maneuverability; L low 、L mid 、L high are low risk, medium risk, and high risk respectively. The UAV maneuverability coefficient is valued according to the inspection requirements, D min is the maximum centripetal acceleration, d min is the minimum safety distance between the UAV and the obstacle. When the UAV is close to the obstacle, or the UAV has a high speed and a small turning radius, then the risk level Level value will also increase accordingly.

[0087] Based on the collision risk assessment results, a risk heat map is generated to visually display the collision risk levels in different areas of the inspection area, and clear visual cues are provided in areas with higher risks to provide safety warnings for path planning and operators.

[0088] Specifically, step 3 constructs a fitness function to evaluate and plan the safe flight path of the UAV, enhancing the flight efficiency and safety of the UAV in a complex environment.

[0089] Define the nearest safe distance from the obstacle as a safety index to measure the collision risk of the path, define the energy consumption of the UAV as a path efficiency index, and construct a multi-objective fitness function based on the above indexes to evaluate the quality of the planned path:

[0090]

[0091] Among them, P is the flight power of the UAV, and the fitness function F fitness is jointly composed of the minimized risk level and the lowest energy consumption.

[0092] Specifically, step 4 solves the optimal path based on the improved particle swarm optimization algorithm, adjusts the flight path according to the simulation results, and ensures that all constraint conditions and fitness criteria are met;

[0093] The path planning in the present invention includes the constraint conditions composed of the UAV and the environment and the minimum fitness value function to be pursued.

[0094] The constructed environmental model is used as the inspection area simulation for the UAV path planning. In the environmental model, random dynamic obstacles are added, and the visual perception module (RGB binocular camera) carried by the UAV is used to detect new obstacles in the environment and perform ranging, edge detection, and model update online.

[0095] Based on the particle swarm optimization algorithm and the maneuverability of the UAV, it helps the UAV avoid obstacles, select a suitable flight route, and evaluate the quality of the path planned by the algorithm according to the value of the fitness function and the iteration time of the algorithm, and optimize the flight strategy.

[0096] According to the complexity of the inspection area, the initial population size and particle positions are set, and each particle represents a possible flight path. Based on the local optimal solution (personal best position) and the global optimal solution (group best position), the search direction of the particle swarm is guided. According to the number of iterations and the current optimization situation, the inertia weight and cognitive factor are adaptively adjusted to balance the global search and local search capabilities, and the speed and position of each particle are updated.

[0097] The improved particle swarm optimization algorithm specifically includes:

[0098] S41: Initialize parameters such as the population size, maximum number of iterations, inertia weight, and cognitive factor, as well as the particle velocity and position;

[0099] S42: The particle calculates the fitness value according to the constructed fitness function; the fitness value reflects the quality of the particle in the planning model, and the goal is to find the solution that minimizes the fitness value.

[0100] S43: Update the inertia weight and cognitive factor according to the improved inertia weight and cognitive factor formulas; at the same time, a perturbation factor is introduced to enhance the reliability of the particle. According to the velocity and the improved position formula, update the velocity and position of the particle;

[0101] S44: Update the individual best and global best: Compare the current fitness value of the particle with its historical best value. If the current value is better, update the individual best position to the current position. Among the individual best positions of all particles, determine the global best position and find the individual best position of the particle with the best fitness value as the global best position;

[0102] S45: Judge the termination condition. If the maximum number of iterations is reached, the algorithm terminates and outputs the global best position and fitness value; otherwise, return to S42 to continue the iteration.

[0103] During the overall iteration process of the algorithm, to maximize the excitation of the algorithm's computing power and the center of gravity direction, increase the inertia weight in the early stage of iteration for global extreme value search. In the later stage of iteration, to search for more accurate extreme values, the inertia weight should be reduced to enhance the local search ability.

[0104] The inertia weight considering the individual extreme value and global extreme value in the iteration progress has strong extreme value search ability and convergence:

[0105]

[0106] Among them, is the inertia weight, are the set maximum and minimum inertia weights respectively, k is the k-th iteration, is the individual best value of the k-th iteration, G best is the group best value, is the average value of the individual best values.

[0107] The cognitive factors in the algorithm include individual cognition and social cognition. The value of individual cognition affects the degree of attention of the particle to the individual best value found in the past while social cognition reflects the degree of attention of the particle to the group experience. Therefore, to accelerate the particle diversity and global exploration ability in the initial stage of the algorithm and enhance the convergence ability in the area near the particle's own position in the later stage, the following improved cognitive factors are adopted in combination with the number of iterations:

[0108] c1 = c1 max -e kK 1 + e kK

[0109] c2 = c2 min +e kK 1 + e kK

[0110] Among them, c1 and c2 are the individual cognitive factor and social cognitive factor respectively, k is the current number of iterations, K is the total number of iterations of the algorithm, and the natural constant e is a fixed constant, approximately 2.71828.

[0111] Through multiple iterations, the algorithm continuously updates the positions of the particles to find the position of the minimum value of the fitness function. A random perturbation method is adopted to improve the global optimization ability of the particles:

[0112]

[0113] Among them, V i k+1 、 is the velocity and position of particle i at the (k + 1)-th iteration, r1 and r2 are random numbers uniformly distributed in the interval, and Δx is the random perturbation factor.

[0114] The present invention improves the problems of convergence speed and falling into local of the particle swarm itself, and then adjusts the weights and cognitive factors that affect the iteration speed and direction of the algorithm, and introduces a perturbation factor to enhance the robustness of the algorithm.

[0115] Output the optimal flight path obtained through optimization, including detailed parameter information such as flight route, flight duration, and flight pacing, and send the optimal path to the UAV control system to guide the UAV to perform inspection requirements according to the planned path. During the execution of the task by the UAV, continuously monitor the flight state and adjust the flight path according to the actual situation.

[0116] Regarding the content disclosed in this case, the following points need to be explained:

[0117] (1). The attached drawings of the embodiments disclosed in this case only relate to the structures involved in the embodiments disclosed in this case, and other structures can refer to the general design;

[0118] (2). Without conflict, the embodiments disclosed in this case and the features in the embodiments can be combined with each other to obtain new embodiments;

[0119] The above is only the specific implementation manner disclosed in this case, but the protection scope of the present disclosure is not limited thereto. The protection scope disclosed in this case shall be subject to the protection scope of the claims.

Claims

1. An autonomous path planning method for an unmanned aerial vehicle based on a visual perception module, characterized in that, It includes the following steps: Step 1: Collect the coordinates and image features of inspection targets and obstacles in the inspection area, and build an environmental model of the UAV inspection area; Step 2: Based on the environmental model, evaluate the collision risk level of the obstacle area according to the maneuverability of the UAV; Step 3: Based on the risk level, establish a fitness function for the safe planned path of the UAV; Step 4: Based on the improved particle swarm optimization algorithm, optimize the fitness function to obtain the optimal planned path for the current inspection area; In Step 2, according to the maneuverability of the UAV, evaluate the maneuverability of the UAV, and then combine with the environmental model to evaluate the collision risk level of the obstacle area; Collect the basic performance parameters of the inspection UAV, including the maximum flight speed, maximum acceleration, minimum turning radius, maximum climb rate and descent rate, wherein, In the formula, is the flight speed of the UAV in the t period, and v max is the maximum speed of the UAV flight, is the flight speed of the UAV in the t + Δt period, is the acceleration of the UAV in the Δt period; is the turning radius of the drone during the t period, r min is the minimum turning radius of the drone are the climb rate and descent rate of the drone during the Δt period respectively is the flight altitude of the drone at the t + Δt period is the flight altitude of the drone at the t period Analyze the maneuvering range of the UAV in different flight modes such as hovering, straight flight, and turning, and evaluate the maneuverability according to the maneuvering range: wherein, 1, 2, and 3 are the levels of maneuverability; The environmental features in the inspection area are sequentially modeled as similar mathematical models according to the edge features, and the minimum safety distance d between the drone and any point in the three-dimensional space is calculated t , which is expressed as: Among them, is the coordinate of the center point of the drone; (x, y, z) is any point on the plane; R is the radius size when the drone is modeled as a circle; Combine the speed, turning radius of the UAV at time t, and the minimum safety distance from the obstacle to evaluate the risk level of the flight area; wherein, Among them, Level is the risk level, L low 、L mid 、L high are low risk, medium risk, and high risk respectively, D min is the maximum centripetal acceleration, d min is the minimum safety distance between the drone and the obstacle; In Step 3, the fitness function is: Among them, F fitness is the fitness function, and P is the flight power of the UAV.

2. The UAV autonomous path planning method based on a visual perception module according to claim 1, characterized in that, In Step 1, use the UAV to carry a high-resolution binocular camera to perform an omnidirectional scan of the inspection area to obtain high-definition image and video data; at the same time, use image processing technology for edge detection to extract the feature information of inspection targets and obstacles from the images, including shape and size; Use GPS positioning technology to record the flight position coordinates of the UAV, the coordinates of inspection targets, and the geographical location coordinates of obstacles; Utilize the collected image and coordinate data, and adopt positioning and mapping technology to combine the coordinates with the image features to form an environmental model.

3. The UAV autonomous path planning method based on a visual perception module according to claim 1, characterized in that, In Step 1, it further includes: after normalizing and cross-validating the collected image and coordinate data, establish a data indexing and retrieval mechanism, and store the relevant data in the inspection database.

4. The UAV autonomous path planning method based on a visual perception module according to claim 1, characterized in that, In Step 4, use the constructed environmental model as a simulation of the inspection area for UAV path planning. In the environmental model, add random dynamic obstacles, use the visual perception module carried by the UAV to detect new obstacles in the environment, and perform ranging, edge detection, and model update online; Based on the particle swarm optimization algorithm and the maneuverability of the UAV, help the UAV avoid obstacles, select a suitable flight route, and evaluate the quality of the path planned by the algorithm according to the value of the fitness function and the time of algorithm iteration, and optimize the flight strategy.

5. The UAV autonomous path planning method based on a visual perception module according to claim 4, characterized in that, According to the complexity of the inspection area, set the initial population size and particle positions, and each particle represents a possible flight path; Based on the local optimal solution and the global optimal solution, guide the search direction of the particle swarm; According to the number of iterations and the current optimization situation, adaptively adjust the inertia weight and the cognitive factor to balance the global search and local search capabilities, and update the velocity and position of each particle.

6. The unmanned aerial vehicle autonomous path planning method based on a visual perception module according to claim 5, wherein Step 4 includes: S41: Initialize the population size, the maximum number of iterations, the inertia weight and the cognitive factor, as well as the particle velocity and position; S42: The particle calculates the fitness value according to the constructed fitness function; S43: Update the inertia weight and the cognitive factor according to the improved inertia weight and cognitive factor formulas; at the same time, introduce a perturbation factor; update the particle velocity and position according to the velocity and the improved position formula; S44: Update the individual optimum and the global optimum: Compare the current fitness value of the particle with the individual historical optimum value. If the current value is better, update the individual optimum position to the current position; among the individual optimum positions of all particles, determine the global optimum position, and find the individual optimum position of the particle with the best fitness value as the global optimum position; S45: Judge the termination condition. If the maximum number of iterations is reached, the algorithm terminates, and the global optimum position and the fitness value are output; otherwise, return to S42 to continue the iteration.

Images