Optimal path planning system for UAV based on improved ant colony algorithm

Abstract

The present invention discloses an optimal path planning system for unmanned aerial vehicles based on an improved ant colony algorithm. The present invention generates accessible points and inaccessible points by identifying the highest obstacle height in a grid and comparing it with the preset maximum allowable flight height of the unmanned aerial vehicle, and performs flight distance cost and flight direction cost analysis on any two adjacent grids. When calculating the flight distance cost, the impact of obstacles on energy consumption is taken into account. By calculating the Euclidean distance and adjusting the cost function according to the presence or absence of obstacles, the path planning is more in line with the actual energy consumption requirements. At the same time, the flight direction cost function is introduced to quantify the turning cost of the unmanned aerial vehicle, and the flight distance and direction change cost are comprehensively considered, so that a better path can be planned, avoiding the non-optimal path problem caused by ignoring the turning cost, solving the problem of ignoring the additional cost caused by the turning process of the unmanned aerial vehicle in the prior art, and improving the accuracy and comprehensiveness of the path planning.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) path planning, and in particular to an optimal path planning system for UAV based on an improved ant colony algorithm. Background Art

[0002] In recent years, with the continuous decline in the manufacturing cost of drones and the gradual increase in the types of functions, drone technology has played an increasingly important role in various fields such as remote monitoring of power lines, pipelines and other infrastructure, climate monitoring, radio communications, etc. Due to the expansion of drone application fields, the variability of operating environments and the complexity of flight missions, drone system accidents occur frequently. Therefore, the problem of drone full coverage path planning has correspondingly become a major research hotspot.

[0003] However, the optimal path planning system for UAVs in the prior art still has the following deficiencies in practical applications: There are still many limitations when using ant colony algorithms to plan the paths of drones. The search time is long, especially when facing large-scale ant colonies. The algorithm needs to screen out better paths from a large number of chaotic paths. This process is often time-consuming and labor-intensive. Moreover, when evaluating the quality of paths, traditional ant colony algorithms only consider the length of the flight distance and ignore the additional costs caused by the drone's turning process. In fact, when turning, drones will consume more energy due to factors such as aerodynamics, and frequent turns will also reduce flight efficiency.

[0004] To this end, an optimal path planning system for UAVs based on an improved ant colony algorithm was introduced. Summary of the invention

[0005] In view of this, the present invention provides an optimal path planning system for a UAV based on an improved ant colony algorithm to solve the problems raised by the above background technology.

[0006] The purpose of the present invention can be achieved by the following technical solution: an optimal path planning system for unmanned aerial vehicles based on an improved ant colony algorithm, comprising: Environmental modeling module: converts flight environment information into raster data, models the UAV flight environment, and generates accessible and inaccessible points of the UAV corresponding to the current flight environment information based on the results of modeling comparison; Cost calculation module: obtain the UAV flight environment modeling, analyze the flight distance cost and flight direction cost of any two adjacent grids in the modeling, and thus obtain the UAV flight distance cost function between any two adjacent grids and flight direction cost function D(x, y); Improved algorithm module: executes the improved ant colony algorithm for path planning, selects the optimal path as the flight path of the UAV based on the result of path planning and sends it to the control execution module; Control execution module: receives the flight path of the drone and controls the drone to fly according to the received flight path; Result evaluation module: After the UAV finishes flying according to the received flight path, the efficiency performance parameters of the UAV at the end of the current flight path are collected and comprehensively evaluated to obtain the efficiency evaluation index xrt corresponding to the current flight path, and the corresponding steps are executed based on the efficiency evaluation index xrt corresponding to the current flight path; the efficiency performance parameters include flight power consumption and flight time.

[0007] In some embodiments, based on the results of the modeling comparison, accessible points and inaccessible points of the drone corresponding to the current flight environment information are generated, specifically: According to the actual terrain conditions of the flight environment, the flight environment is divided into several 10×10m cubic grids. The height of the highest obstacle in the grid is abstracted as the height of the grid cube, and each grid is represented in the form of coordinate points (x, y, z). The maximum allowed flight altitude of the drone is preset; the height of each grid is compared with the preset maximum allowed flight altitude. If the height of a grid is less than or equal to the preset maximum allowed flight altitude, the grid is marked as a drone accessible point, otherwise it is marked as a drone inaccessible point; it is expressed as , where o represents the number of each grid, and h represents the preset maximum allowed flight altitude.

[0008] In some embodiments, the cost function of the flight distance of the drone between any two adjacent grids is obtained. , specifically: For any two adjacent grid nodes and , calculate the Euclidean distance between them After the calculation is completed, for any two adjacent grid nodes, identify whether there is an obstacle, that is, compare the highest obstacle height in the two sets of adjacent grids with the maximum allowed flight height. If it is greater than the maximum allowed flight height, it is determined that there is an obstacle on the ij path, otherwise it is determined that there is no obstacle on the ij path; When there are obstacles on the ij path, the flight distance cost function between any two adjacent grids ;in represents the highest obstacle height between any two adjacent grids; if there is no obstacle on the ij path, the flight distance cost function = .

[0009] In some embodiments, the flight direction cost function D(x, y) of the drone in any two adjacent grids is obtained as follows: Set the angle between the straight line formed by node j and the current node i and the horizontal coordinate to , the angle between the current node i and the last covered node is ,for and Calculate the difference and take the absolute value to get the change in direction of the drone when it moves from the current node i to the next node j ; According to the change value Calculate the flight direction cost function D(x, y), that is, through the formula Calculate and obtain.

[0010] In some embodiments, the improved ant colony algorithm is executed to perform path planning, specifically: M1: Initialize the ant colony size v, where v is the number of ants, where v>40, and randomly place each ant at the starting point; M2: Using the formula Initialize the pheromone of each path ij; when the ant passes through the path ij, according to the formula Update Path Pheromones ; Where Q represents the total amount of pheromone set in advance; M3: For each ant, according to the transfer probability formula of the improved ant colony algorithm Calculate the transition probability at time t ;in is the preset information heuristic index, b is the preset expected heuristic index, Indicates the pheromone value between nodes at that moment; represents the set of all nodes that the ant numbered k can visit after starting from node i; It represents the heuristic factor value between the nodes at that moment, that is, the inverse of the Euclidean distance between the nodes at that moment; M4: Transition probability according to step M3 Combined with the roulette algorithm, the next node to be visited is selected until each ant reaches the target point. When v ants complete a path construction, the performance evaluation index frea of ​​each path is analyzed, and the path with the largest performance evaluation index frea is selected as the candidate path.

[0011] In some embodiments, the optimal path is selected as the flight path of the drone based on the result of path planning, specifically: M5: Preset an initial mutation probability and randomly select two nodes to exchange; update the pheromone between all path nodes, using After the update is completed, the performance evaluation index frea of ​​each path is analyzed again; M6: After repeating step M5 for a set number of times, the path with the largest performance evaluation index frea is selected as the optimal path.

[0012] In some embodiments, the performance evaluation index frea of ​​each path is analyzed, specifically: M4-401: Accumulate the flight distance cost functions between nodes on each path as the distance cost value corresponding to each path; accumulate the flight direction cost functions between nodes on each path as the direction cost value corresponding to each path; and count the number of inaccessible points that the drone needs to pass through on each path as the number of anomalies corresponding to each path; M4-402: Mark the distance cost, direction cost, and number of anomalies of each path as Sa, Sb, and Sc, respectively, and substitute them into the formula A weighted calculation is performed to obtain the performance evaluation index frea of ​​each path; wherein Za, Zb and Zc are the preset maximum allowable distance cost value, the maximum allowable direction cost value and the maximum allowable number of anomalies respectively; m1, m2 and m3 are the influence weight factors of the distance cost value Sa, the direction cost value Sb and the number of anomalies Sc respectively.

[0013] In some embodiments, the efficiency evaluation index xrt corresponding to the current flight path is obtained, specifically: After receiving the flight path of the drone, according to the planned flight path, the reference maximum values ​​of the drone's flight power consumption and flight duration corresponding to the current flight path are preset, that is, the reference maximum flight power consumption and the reference maximum flight duration; The power consumption and flight duration of the drone are marked as Ra and Rb respectively, and then substituted into the formula Perform weighted calculation to obtain the efficiency evaluation index xrt of the UAV corresponding to the current flight path; and They represent the maximum reference power consumption and maximum reference flight duration respectively; and They are the influencing weight factors of flight power consumption Ra and flight duration Rb respectively.

[0014] In some embodiments, corresponding steps are performed based on the efficiency evaluation index xrt corresponding to the current flight path, specifically: Compare the efficiency evaluation index xrt of the current flight path with the preset efficiency threshold index. If the efficiency evaluation index xrt of the current flight path is less than the preset efficiency threshold index, trigger the traceback signaling; Taking the time point of triggering the traceback signal as the starting point, extract the historical efficiency evaluation indexes of each group of drones in the set time period before the starting point, and integrate them with the current efficiency evaluation index as the evaluation data packet of the ant colony algorithm in this time interval; The number of times the traceback signaling is triggered in the evaluation data packet is counted as the number of performance deviations, and the efficiency evaluation index corresponding to each triggering of the traceback signaling is extracted. The difference between each group of extracted efficiency evaluation index and the preset efficiency threshold index is calculated, and the absolute value of the calculated difference is taken and the average is obtained as the performance gap value of the ant colony algorithm in the time interval; Preset the weight coefficients corresponding to the number of performance deviations and the performance gap value, respectively, multiply the number of performance deviations and the performance gap value by the corresponding weight coefficients, and then sum them up to obtain the algorithm performance index of the ant colony algorithm in the time interval; The obtained algorithm performance index and evaluation data package are sent to the technical staff. After receiving the algorithm performance index and evaluation data package, the technical staff will analyze and evaluate them. If the evaluation result shows that it is not affected by environmental factors, the ant colony algorithm will be adjusted and optimized.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention generates accessible points and inaccessible points by identifying the highest obstacle height in the grid and comparing it with the preset maximum allowed flight height of the drone, and performs flight distance cost and flight direction cost analysis on any two adjacent grids. When calculating the flight distance cost, the impact of obstacles on energy consumption is taken into account. Through Euclidean distance calculation and adjusting the cost function according to the presence or absence of obstacles, the path planning is more in line with the actual energy consumption requirements. At the same time, the flight direction cost function is introduced to quantify the drone turning cost, and the flight distance and direction change cost are comprehensively considered, so that a better path can be planned, avoiding the non-optimal path problem caused by ignoring the turning cost, solving the problem of ignoring the additional cost caused by the turning process of the drone in the prior art, and improving the accuracy and comprehensiveness of path planning; The present invention guides ants to choose a better path and improves the time efficiency of path planning by updating pheromones according to a formula including directional cost when ants pass through the path. The invention also introduces mutation operators in combination with genetic algorithms, presets initial mutation probabilities and randomly exchanges nodes. If the new path is better, it is retained, thereby enhancing the algorithm's optimization ability and avoiding falling into a local optimal solution, so that the algorithm can more effectively search for the optimal path in a complex solution space. The closed-loop feedback mechanism constructed by the result evaluation module of the present invention enables the algorithm to be continuously improved according to the actual flight data. Over time, the algorithm can better adapt to different environments and tasks, and its stability and performance are continuously improved. For example, after multiple flight missions in different seasons, different terrains, and different mission types, the algorithm can be gradually optimized and adapted to various complex situations, providing strong support for the long-term stable and efficient operation of drones. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Further details, features and advantages of the present application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 It is a principle block diagram of the present invention; Figure 2 It is a schematic diagram of the ant colony algorithm process in the present invention. DETAILED DESCRIPTION

[0017] Several embodiments of the present application will be described in more detail below with reference to the accompanying drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and to fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0018] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification, and will not be interpreted in an idealized or overly formal sense unless explicitly defined as such herein.

[0019] See also Figure 1-Figure 2 As shown, the optimal path planning system for UAV based on the improved ant colony algorithm includes an environment modeling module, a cost calculation module, an improved algorithm module, a control execution module and a result evaluation module; The environment modeling module is used to convert the flight environment information into raster data using the raster method, and to model the UAV flight environment. Based on the results of the modeling comparison, the accessible points and inaccessible points of the UAV corresponding to the current flight environment information are generated; Specifically: According to the actual terrain conditions of the flight environment, the flight environment is divided into several 10×10m cubic grids. The height of the highest obstacle in the grid is abstracted as the height of the grid cube, and each grid is represented in the form of coordinate points (x, y, z). According to the flight mission type of the drone in the current flight environment, the maximum allowable flight altitude of the drone is preset; the height of each grid is compared with the preset maximum allowable flight altitude. If the height of a grid is less than or equal to the preset maximum allowable flight altitude, the grid is marked as a drone accessible point, otherwise it is marked as a drone inaccessible point; it is expressed as , where o represents the number of each grid, and h represents the preset maximum allowed flight altitude; The maximum allowed flight altitude is the initial altitude set for the drone; The cost calculation module is used to obtain the UAV flight environment modeling. In the modeling, the flight distance cost and flight direction cost of any two adjacent grids are analyzed respectively, so as to obtain the flight distance cost function of the UAV in any two adjacent grids. and flight direction cost function D(x, y); Specifically: For any two adjacent grid nodes and , calculate the Euclidean distance between them After the calculation is completed, for any two adjacent grid nodes, identify whether there is an obstacle, that is, compare the highest obstacle height in the two sets of adjacent grids with the maximum allowed flight height. If it is greater than the maximum allowed flight height, it is determined that there is an obstacle on the ij path, otherwise it is determined that there is no obstacle on the ij path; The highest obstacle heights in the two groups of grids are obtained by analyzing the obstacle height information in the grid data; When there are obstacles on the ij path, the flight distance cost function between any two adjacent grids ;in Represents the highest obstacle height between any two adjacent grids; If there are no obstacles on the ij path, the flight distance cost function = ; It should be noted that, through the above calculations, the increased energy consumption of the drone during obstacle-crossing flight is taken into account in path planning, making the path planning more in line with the actual flight energy consumption requirements; Set the angle between the straight line formed by node j and the current node i and the horizontal coordinate to , the angle between the current node i and the last covered node is ,for and Calculate the difference and take the absolute value to get the change in direction of the drone when it moves from the current node i to the next node j ; According to the change value Calculate the flight direction cost function D(x, y), that is, through the formula Calculate and obtain; It should be noted that the larger the flight direction cost function D(x, y), the greater the direction change of the drone, and the greater the cost of the corresponding direction change. The direction cost function is introduced to comprehensively consider the cost of flight distance and direction change in path planning, so as to make the planned path better.

[0020] The improved algorithm module is used to execute the improved ant colony algorithm for path planning. Based on the result of path planning, the optimal path is selected as the flight path of the UAV and sent to the control execution module; Specifically: M1: Initialize the ant colony size v, where v is the number of ants, where v>40. The specific number is initialized and set by the technicians; and randomly place each ant at the starting point; M2: Using the formula Initialize the pheromone of each path ij; when the ant passes through the path ij, according to the formula Update Path Pheromones ; Where Q represents the total amount of pheromone set in advance; It should be noted that this initialization method can speed up the start of the positive feedback mechanism in the early stage of the algorithm, making the ants tend to choose a better path more quickly, thereby shortening the algorithm convergence time; By introducing the direction cost, the pheromone update not only considers the path length, but also the direction change factor, which guides the ants to choose a better path in subsequent searches and improves the time efficiency of path planning. M3: For each ant, according to the transfer probability formula of the improved ant colony algorithm Calculate the transition probability at time t ;in is the preset information heuristic index, b is the preset expected heuristic index, Indicates the pheromone value between nodes at that moment; represents the set of all nodes that the ant numbered k can visit after starting from node i; It represents the heuristic factor value between the nodes at that moment, that is, the inverse of the Euclidean distance between the nodes at that moment; M4: Transition probability according to step M3 Combined with the roulette algorithm, the next node to be visited is selected until each ant reaches the target point. When v ants complete a path construction, the performance evaluation index frea of ​​each path is analyzed, and the path with the largest performance evaluation index frea is selected as the candidate path; M4-401: Accumulate the flight distance cost functions between nodes on each path as the distance cost value corresponding to each path; accumulate the flight direction cost functions between nodes on each path as the direction cost value corresponding to each path; and count the number of inaccessible points that the drone needs to pass through on each path as the number of anomalies corresponding to each path; M4-402: Mark the distance cost, direction cost, and number of anomalies of each path as Sa, Sb, and Sc, respectively, and substitute them into the formula Perform weighted calculation to obtain the performance evaluation index frea of ​​each path; Za, Zb and Zc are respectively the preset maximum allowable distance cost value, the maximum allowable direction cost value and the maximum allowable number of anomalies; m1, m2 and m3 are respectively the influence weight factors of the distance cost value Sa, the direction cost value Sb and the number of anomalies Sc; M5: Preset an initial mutation probability and randomly select two nodes to exchange; the initial mutation probability is 0.1 for example; update the pheromone between all path nodes, using After the update is completed, the performance evaluation index frea of ​​each path is analyzed again; M6: After repeating step M5 for a set number of times, the path with the largest performance evaluation index frea among the candidate paths is selected as the optimal path; The control execution module is used to receive the flight path of the UAV and control the UAV to fly according to the received flight path; It should be noted that during the flight, the flight path is fine-tuned based on the real-time location information and environmental information of the drone, such as through onboard sensors (such as local obstacle avoidance when encountering sudden obstacles), to ensure that the drone completes its flight mission safely and efficiently and achieves the optimal path flight of the drone in complex environments. At the same time, after the flight, the flight data (such as actual flight path, energy consumption, flight time, etc.) can be recorded and analyzed to provide a reference for subsequent path planning and algorithm improvement.

[0021] The result evaluation module is used to collect and comprehensively evaluate the efficiency performance parameters of the UAV after the UAV completes the flight according to the received flight path, so as to obtain the efficiency evaluation index xrt corresponding to the current flight path; the efficiency performance parameters include the flight power consumption and the flight time; and the corresponding steps are executed based on the efficiency evaluation index xrt corresponding to the current flight path; Specifically: After receiving the flight path of the drone, according to the planned flight path, the reference maximum values ​​of the drone's flight power consumption and flight duration corresponding to the current flight path are preset, that is, the reference maximum flight power consumption and the reference maximum flight duration; The power consumption and flight duration of the drone are marked as Ra and Rb respectively, and then substituted into the formula Perform weighted calculation to obtain the efficiency evaluation index xrt of the UAV corresponding to the current flight path; and They represent the maximum reference power consumption and maximum reference flight duration respectively; and are the influencing weight factors of flight power consumption Ra and flight duration Rb respectively; Compare the efficiency evaluation index xrt of the current flight path with the preset efficiency threshold index. If the efficiency evaluation index xrt of the current flight path is less than the preset efficiency threshold index, trigger the traceback signaling; Taking the time point of triggering the traceback signal as the starting point, extract the historical efficiency evaluation indexes of each group of drones in the set time period before the starting point, and integrate them with the current efficiency evaluation index as the evaluation data packet of the ant colony algorithm in this time interval; The number of times the traceback signaling is triggered in the evaluation data packet is counted as the number of performance deviations, and the efficiency evaluation index corresponding to each triggering of the traceback signaling is extracted. The difference between each group of extracted efficiency evaluation index and the preset efficiency threshold index is calculated, and the absolute value of the calculated difference is taken and the average is obtained as the performance gap value of the ant colony algorithm in the time interval; Preset the weight coefficients corresponding to the number of performance deviations and the performance gap value, respectively, multiply the number of performance deviations and the performance gap value by the corresponding weight coefficients, and then sum them up to obtain the algorithm performance index of the ant colony algorithm in the time interval; The obtained algorithm performance index and evaluation data package are sent to the technicians. After receiving the algorithm performance index and evaluation data package, the technicians will analyze and evaluate them. If the evaluation result shows that it is not affected by environmental factors, the ant colony algorithm will be adjusted and optimized; environmental factors include wind direction, wind speed and weather, which will affect the efficiency performance index of the drone on the flight path; It should be noted that the above-mentioned evaluation and optimization mechanism based on flight data forms a closed-loop feedback system. The evaluation results after each flight mission can be used to improve the algorithm, so that the algorithm can continuously adapt to different flight environments and mission requirements. Over time, the algorithm will continue to improve in efficiency and stability, and better serve the path planning and flight mission execution of drones.

[0022] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. The optimal path planning system for UAV based on improved ant colony algorithm is characterized by: include: Environmental modeling module: converts flight environment information into raster data, models the UAV flight environment, and generates accessible and inaccessible points of the UAV corresponding to the current flight environment information based on the results of modeling comparison; Cost calculation module: obtain the UAV flight environment modeling, analyze the flight distance cost and flight direction cost of any two adjacent grids in the modeling, and thus obtain the UAV flight distance cost function between any two adjacent grids and flight direction cost function D(x, y); Improved algorithm module: executes the improved ant colony algorithm for path planning, selects the optimal path as the flight path of the UAV based on the result of path planning and sends it to the control execution module; Control execution module: receives the flight path of the drone and controls the drone to fly according to the received flight path; Result evaluation module: After the UAV finishes flying according to the received flight path, the efficiency performance parameters of the UAV at the end of the current flight path are collected and comprehensively evaluated to obtain the efficiency evaluation index xrt corresponding to the current flight path, and the corresponding steps are executed based on the efficiency evaluation index xrt corresponding to the current flight path; the efficiency performance parameters include flight power consumption and flight time.

2. The optimal path planning system for unmanned aerial vehicles based on the improved ant colony algorithm according to claim 1 is characterized in that: Based on the results of modeling comparison, the accessible points and inaccessible points of the drone corresponding to the current flight environment information are generated, specifically: According to the actual terrain conditions of the flight environment, the flight environment is divided into several 10×10m cubic grids. The height of the highest obstacle in the grid is abstracted as the height of the grid cube, and each grid is represented in the form of coordinate points (x, y, z). The maximum allowed flight altitude of the drone is preset; the height of each grid is compared with the preset maximum allowed flight altitude. If the height of a grid is less than or equal to the preset maximum allowed flight altitude, the grid is marked as a drone accessible point, otherwise it is marked as a drone inaccessible point; it is expressed as , where o represents the number of each grid, and h represents the preset maximum allowed flight altitude.

3. The optimal path planning system for unmanned aerial vehicles based on the improved ant colony algorithm according to claim 2 is characterized in that: Get the cost function of the flight distance of the drone between any two adjacent grids , specifically: For any two adjacent grid nodes and , calculate the Euclidean distance between them After the calculation is completed, for any two adjacent grid nodes, identify whether there is an obstacle, that is, compare the highest obstacle height in the two sets of adjacent grids with the maximum allowed flight height. If it is greater than the maximum allowed flight height, it is determined that there is an obstacle on the ij path, otherwise it is determined that there is no obstacle on the ij path; When there are obstacles on the ij path, the flight distance cost function between any two adjacent grids ;in represents the highest obstacle height between any two adjacent grids; if there is no obstacle on the ij path, the flight distance cost function = .

4. The optimal path planning system for unmanned aerial vehicles based on the improved ant colony algorithm according to claim 3 is characterized in that: The cost function D(x, y) of the flight direction of the drone in any two adjacent grids is obtained as follows: Set the angle between the straight line formed by node j and the current node i and the horizontal coordinate to , the angle between the current node i and the last covered node is ,for and Calculate the difference and take the absolute value to get the change in direction of the drone when it moves from the current node i to the next node j ; According to the change value Calculate the flight direction cost function D(x, y), that is, through the formula Calculate and obtain.

5. The optimal path planning system for unmanned aerial vehicles based on improved ant colony algorithm according to claim 4 is characterized in that: Execute the improved ant colony algorithm for path planning, specifically: M1: Initialize the ant colony size v, where v is the number of ants, where v>40, and randomly place each ant at the starting point; M2: Using the formula Initialize the pheromone of each path ij; when the ant passes through the path ij, according to the formula Update Path Pheromones ; Where Q represents the total amount of pheromone set in advance; M3: For each ant, according to the transfer probability formula of the improved ant colony algorithm Calculate the transition probability at time t ;in is the preset information heuristic index, b is the preset expected heuristic index, Indicates the pheromone value between nodes at that moment; represents the set of all nodes that the ant numbered k can visit after starting from node i; It represents the heuristic factor value between the nodes at that moment, that is, the inverse of the Euclidean distance between the nodes at that moment; M4: Transition probability according to step M3 Combined with the roulette algorithm, the next node to be visited is selected until each ant reaches the target point. When v ants complete a path construction, the performance evaluation index frea of ​​each path is analyzed, and the path with the largest performance evaluation index frea is selected as the candidate path.

6. The optimal path planning system for unmanned aerial vehicles based on the improved ant colony algorithm according to claim 5 is characterized in that: Based on the results of path planning, the optimal path is selected as the flight path of the UAV, specifically: M5: Preset an initial mutation probability and randomly select two nodes to exchange; update the pheromone between all path nodes, using After the update is completed, the performance evaluation index frea of ​​each path is analyzed again; M6: After repeating step M5 for a set number of times, the path with the largest performance evaluation index frea is selected as the optimal path.

7. The optimal path planning system for unmanned aerial vehicles based on improved ant colony algorithm according to claim 6 is characterized in that: Analyze the performance evaluation index frea of ​​each path, specifically: M4-401: Accumulate the flight distance cost functions between nodes on each path as the distance cost value corresponding to each path; accumulate the flight direction cost functions between nodes on each path as the direction cost value corresponding to each path; and count the number of inaccessible points that the drone needs to pass through on each path as the number of anomalies corresponding to each path; M4-402: Mark the distance cost, direction cost, and number of anomalies of each path as Sa, Sb, and Sc, respectively, and substitute them into the formula A weighted calculation is performed to obtain the performance evaluation index frea of ​​each path; wherein Za, Zb and Zc are the preset maximum allowable distance cost value, the maximum allowable direction cost value and the maximum allowable number of anomalies respectively; m1, m2 and m3 are the influence weight factors of the distance cost value Sa, the direction cost value Sb and the number of anomalies Sc respectively.

8. The optimal path planning system for unmanned aerial vehicles based on improved ant colony algorithm according to claim 7 is characterized in that: The efficiency evaluation index xrt corresponding to the current flight path is obtained, specifically: After receiving the flight path of the drone, according to the planned flight path, the reference maximum values ​​of the drone's flight power consumption and flight duration corresponding to the current flight path are preset, that is, the reference maximum flight power consumption and the reference maximum flight duration; The power consumption and flight duration of the drone are marked as Ra and Rb respectively, and then substituted into the formula Perform weighted calculation to obtain the efficiency evaluation index xrt of the UAV corresponding to the current flight path; and They represent the maximum reference power consumption and maximum reference flight duration respectively; and They are the influencing weight factors of flight power consumption Ra and flight duration Rb respectively.

9. The optimal path planning system for unmanned aerial vehicles based on improved ant colony algorithm according to claim 8, characterized in that: Based on the efficiency evaluation index xrt corresponding to the current flight path, the corresponding steps are executed, specifically: Compare the efficiency evaluation index xrt of the current flight path with the preset efficiency threshold index. If the efficiency evaluation index xrt of the current flight path is less than the preset efficiency threshold index, trigger the traceback signaling; Taking the time point of triggering the traceback signal as the starting point, extract the historical efficiency evaluation indexes of each group of drones in the set time period before the starting point, and integrate them with the current efficiency evaluation index as the evaluation data packet of the ant colony algorithm in this time interval; The number of times the traceback signaling is triggered in the evaluation data packet is counted as the number of performance deviations, and the efficiency evaluation index corresponding to each triggering of the traceback signaling is extracted. The difference between each group of extracted efficiency evaluation index and the preset efficiency threshold index is calculated, and the absolute value of the calculated difference is taken and the average is obtained as the performance gap value of the ant colony algorithm in the time interval; Preset the weight coefficients corresponding to the number of performance deviations and the performance gap value, respectively, multiply the number of performance deviations and the performance gap value by the corresponding weight coefficients, and then sum them up to obtain the algorithm performance index of the ant colony algorithm in the time interval; The obtained algorithm performance index and evaluation data package are sent to the technical staff. After receiving the algorithm performance index and evaluation data package, the technical staff will analyze and evaluate them. If the evaluation result shows that it is not affected by environmental factors, the ant colony algorithm will be adjusted and optimized.

Images