Multi-UAV navigation path planning method and system based on multi-task parameters

Abstract

The present invention relates to the field of path planning technology, and is a multi-task parameter-based multi-UAV navigation path planning method and system, comprising: obtaining a three-dimensional mapping space, determining the current wind speed, current gust vector, current rainfall, and UAV task sequence, obtaining multiple historical fault groups, performing negative pheromone marking on the three-dimensional mapping space based on a preset initial pheromone concentration and the multiple historical fault groups to obtain a negative marking space, performing an optimized ant colony simulation on the j-th task coordinate group based on the current wind speed, current gust vector, current rainfall, a pre-constructed optimization heuristic function, a pre-constructed path coordinate group sequence set, and the negative marking space to obtain an updated coordinate group sequence set, generating multiple optimal paths using the updated coordinate group sequence set, and completing multi-UAV navigation path planning based on the multiple optimal paths. The present invention can improve the accuracy and reliability of path planning and enhance the safety of multiple UAVs when performing tasks simultaneously.

Description

Technical Field

[0001] The present invention relates to the field of path planning technology, and in particular to a multi-task parameter-based multi-UAV navigation path planning method and system. Background Art

[0002] With the increasing demand for intelligent operations, drone systems have been widely used in complex environments such as urban security, smart warehousing, and disaster response. Compared to single drone operations, the coordinated execution of multiple drones can significantly improve response speed and coverage. However, in actual applications, when multiple drones simultaneously perform multiple tasks in the same airspace, they may face challenges such as path conflicts and environmental interference, placing higher demands on the accuracy, reliability, and safety of path planning.

[0003] At present, commonly used drone path planning methods include Dijkstra, RRT and ant colony algorithms, which mainly search for feasible paths in space based on grid maps. Some methods introduce obstacle avoidance strategies to adapt to complex tasks, but they are still mainly based on static scenes and single tasks.

[0004] Although existing technologies can plan drone navigation paths, existing ant colony algorithms generally lack consideration of environmental factors such as wind speed and rainfall, and fail to effectively use historical fault data for pheromone marking to avoid high-risk areas. In addition, there is a lack of interference analysis mechanism between multiple drone paths, which easily leads to path conflicts. Therefore, there is an urgent need for a navigation path planning method for multiple tasks and multiple drones to improve the accuracy and reliability of path planning and enhance the safety of multiple drones performing tasks simultaneously. Summary of the Invention

[0005] The present invention provides a multi-UAV navigation path planning method based on multi-task parameters and a computer-readable storage medium, the main purpose of which is to improve the accuracy and reliability of path planning and enhance the safety of multiple UAVs when performing tasks simultaneously.

[0006] To achieve the above objectives, the present invention provides a multi-UAV navigation path planning method based on multi-task parameters, comprising:

[0007] Acquire a three-dimensional mapping space, wherein the three-dimensional mapping space is in the shape of a cube and includes: a spatial rectangular coordinate system and a plurality of cells, wherein the cell is in the shape of a cube, a unit length of the spatial rectangular coordinate system is a side length of the cell, each cell in the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstruction cells;

[0008] Determine the current wind speed, current gust vector, current rainfall, and the UAV mission sequence, wherein the UAV mission sequence includes: K mission coordinate groups, and the mission coordinate group includes: starting coordinates and end coordinates;

[0009] Acquire multiple historical fault groups, where the historical fault groups include: historical fault coordinates and historical fault times;

[0010] Based on the preset initial pheromone concentration and multiple historical fault groups, the three-dimensional mapping space is negatively pheromone-marked to obtain a negatively marked space;

[0011] Extract the jth task coordinate group from the UAV task sequence, where the initial value of j is 1;

[0012] An ant colony simulation is performed on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative label space to obtain an updated coordinate group sequence set, wherein the updated coordinate group sequence set includes: multiple optimal path sequences;

[0013] Using the updated coordinate group sequence set as the path coordinate group sequence set, let J = j + 1, let J be j, return to the step of extracting the j-th task coordinate group from the UAV task sequence, until j = K, and use the updated coordinate group sequence set to generate multiple optimal paths, where the optimal paths correspond one-to-one to the task coordinate groups;

[0014] Complete multi-UAV navigation path planning based on multiple optimal paths.

[0015] Optionally, performing negative pheromone marking on the three-dimensional mapping space based on a preset initial pheromone concentration and a plurality of historical fault groups to obtain a negatively marked space includes:

[0016] Perform the following operations on each of the multiple historical fault groups:

[0017] Identify the fault cells in the three-dimensional mapping space based on the historical fault coordinates in the historical fault group;

[0018] Identifying a plurality of dangerous cells in a three-dimensional mapping space based on the faulty cell, wherein the dangerous cells are adjacent to the faulty cell in the three-dimensional mapping space;

[0019] Calculate the danger pheromone concentration based on the initial pheromone concentration and the number of historical failures in the historical failure group;

[0020] Calculate the barrier pheromone concentration based on the danger pheromone concentration, where the barrier pheromone concentration is half of the danger pheromone concentration;

[0021] Marking a fault cell based on the obstacle pheromone concentration to obtain an obstacle-marked cell, and marking each of a plurality of dangerous cells based on the danger pheromone concentration to obtain a plurality of danger-marked cells;

[0022] Aggregating the obstacle marking cells and the plurality of danger marking cells to obtain a plurality of exclusion cells, wherein the exclusion cells are obstacle marking cells or danger marking cells;

[0023] Eliminate multiple exclusion cells from multiple cells in the three-dimensional mapping space to obtain multiple basic cells;

[0024] Each of the plurality of basic cells is marked based on the initial pheromone concentration to obtain a plurality of initially marked cells;

[0025] Aggregating multiple exclusion cells and multiple initial mark cells to obtain multiple initial cells, wherein the initial cells are exclusion cells or initial mark cells;

[0026] A negative label space is identified based on multiple initial cells and a three-dimensional mapping space.

[0027] Optionally, the optimizing ant colony simulation is performed on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set, and the negative label space to obtain the updated coordinate group sequence set, including:

[0028] According to the starting coordinates in the j-th task coordinate group, the starting cell is identified in the negative marking space, and the ending coordinates in the j-th task coordinate group are marked as the ending coordinates;

[0029] Generate multiple virtual ants, and perform the following operations on each of the multiple virtual ants:

[0030] Place the virtual ant into the starting cell to obtain the starting cell;

[0031] Identify multiple neighboring cells of the starting cell in the negatively marked space;

[0032] For each of multiple neighboring cells, perform the following operations:

[0033] Perform transfer strategy analysis on neighboring cells and starting cells based on current wind speed, current gust vector, current rainfall, optimization heuristic function, termination coordinates, preset number of loops, pre-built memory coordinate group sequence, preset start time and path coordinate group sequence set to obtain transfer comprehensive index and update time;

[0034] Summarize the transfer comprehensive indexes and update times respectively to obtain multiple transfer comprehensive indexes and multiple update times, wherein the transfer comprehensive indexes and update times are in one-to-one correspondence with adjacent cells;

[0035] Based on multiple transfer comprehensive indexes and multiple update times, multiple adjacent cells are probabilistically transferred to obtain the transferred cells and landing time;

[0036] Move the virtual ant to the transfer cell, obtain the landing cell, and identify the landing coordinates of the landing cell in the negative mark space;

[0037] Combine the landing point coordinates and the landing time to obtain a landing point coordinate group;

[0038] Using the landing point coordinate group to update the memory coordinate group sequence, to obtain an updated memory group sequence;

[0039] Determine whether the landing point coordinates are equal to the ending coordinates;

[0040] If the landing point coordinates are not equal to the ending coordinates, the landing point cell is used as the starting cell, the updated memory group sequence is used as the memory coordinate group sequence, the landing time is used as the starting time, and the process returns to the step of identifying multiple neighboring cells of the starting cell in the negative mark space until the landing point coordinates are equal to the ending coordinates;

[0041] If the landing point coordinates are equal to the ending coordinates, the updated memory group sequence is used as the target path sequence;

[0042] Summarizing the target path sequences to obtain multiple target path sequences, wherein the target path sequences include: multiple target coordinate groups, and the target coordinate groups include: target coordinates and target time;

[0043] Perform path analysis on multiple target path sequences to obtain the path overlap and the total time of multiple paths;

[0044] Determine the path overlap with a preset path threshold and compare the number of cycles with a preset cycle threshold;

[0045] If the path overlap is less than the path threshold and the number of cycles is less than the cycle threshold, pheromone optimization is performed on the negative marking space according to the total time of multiple paths and the multiple target path sequences to obtain a path marking space, the number of updates is calculated according to the number of cycles, the path marking space is used as the negative marking space, the number of updates is used as the number of cycles, and the step of generating multiple virtual ants is returned to until the path overlap is greater than or equal to the path threshold or the number of cycles is greater than or equal to the cycle threshold;

[0046] If the path overlap is greater than or equal to the path threshold or the number of cycles is greater than or equal to the cycle threshold, the optimal path sequence is determined based on the total time of multiple paths and multiple target path sequences, and the path coordinate group sequence set is updated according to the optimal path sequence to obtain an updated coordinate group sequence set.

[0047] Optionally, the transfer strategy analysis is performed on the neighboring cells and the starting cell according to the current wind speed, the current gust vector, the current rainfall, the optimization heuristic function, the end coordinates, the preset number of cycles, the pre-built memory coordinate group sequence, the preset start time and the path coordinate group sequence set to obtain the transfer comprehensive index and update time, including:

[0048] Identify the neighboring coordinates and starting coordinates of the neighboring cells and starting cell in the negative label space respectively;

[0049] Calculate the far distance based on the adjacent coordinates and the ending coordinates;

[0050] Get the nearest distance based on the neighboring coordinates and the starting coordinates;

[0051] Calculate the moving time based on the preset resolution conversion rate, the nearest distance and the preset average speed;

[0052] Calculate the update time based on the moving time and the starting time, where the update time is the sum of the moving time and the starting time;

[0053] Determine whether the adjacent cell is the obstacle cell, and if the adjacent cell is the obstacle cell, use the preset zero probability as the transfer comprehensive index;

[0054] Otherwise, the movement vector is determined based on the neighboring coordinates and the starting coordinates;

[0055] Identifying a last coordinate group in the memory coordinate group sequence, wherein the last coordinate group is the last memory coordinate group in the memory coordinate group sequence;

[0056] Identifying an advance coordinate group in the memory coordinate group sequence based on the last coordinate group, wherein the last coordinate group is adjacent to the advance coordinate group in the memory coordinate group sequence and lags behind the advance coordinate group;

[0057] Extracting an advance coordinate from the advance coordinate group, extracting a final coordinate from the final coordinate group, obtaining a memory vector based on the advance coordinate and the final coordinate, and calculating a corner energy consumption index based on the memory vector and the motion vector;

[0058] The environmental impact index is calculated based on the movement vector, current wind speed, current gust vector, and current rainfall. The calculation formula is as follows:

[0059]

[0060] in, is the environmental impact index, and are the current rainfall and current wind speed, respectively. and are the current gust vector and the preset rise vector respectively, is the movement vector, is the modulus of the motion vector, and are the preset lift resistance coefficient and the preset wind speed resistance coefficient, and are the modulus of the lift vector and the modulus of the current gust vector respectively;

[0061] Identify neighboring pheromone concentrations of neighboring cells;

[0062] Based on the update time and adjacent coordinates, the interference characteristics of the path coordinate group sequence set are analyzed to obtain the mutual interference index;

[0063] The number of cycles, neighboring pheromone concentration, near distance, far distance, mutual interference index, corner energy consumption index and environmental impact index are substituted into the optimization heuristic function to obtain the transfer comprehensive index.

[0064] Optionally, performing interference feature analysis on the path coordinate group sequence set based on the update time and the adjacent coordinates to obtain a mutual interference index includes:

[0065] Determine whether the path coordinate group sequence set is an empty set. If the path coordinate group sequence set is an empty set, use the preset initial interference index as the mutual interference index. Otherwise, perform the following operations on each path coordinate group sequence in the path coordinate group sequence set:

[0066] identifying close coordinate groups in the path coordinate group sequence based on update time;

[0067] Extracting close coordinates and close time from the close coordinate group;

[0068] Calculate the mutual interference factor based on the update time, adjacent coordinates, close coordinates and close time;

[0069] The mutual interference factors are aggregated to obtain multiple mutual interference factors, and a mutual interference index is calculated according to the multiple mutual interference factors, wherein the mutual interference index is the sum of the multiple mutual interference factors.

[0070] Optionally, the optimization heuristic function is as follows:

[0071]

[0072] in, To optimize the heuristic function, is the corner energy consumption index, is the mutual interference index, is the neighboring pheromone concentration, is the number of cycles, is the cycle threshold, For the distance, For close distance.

[0073] Optionally, performing probabilistic transfer of multiple adjacent cells based on multiple transfer comprehensive indexes and multiple update times to obtain transfer cells and landing times includes:

[0074] For each of multiple neighboring cells, perform the following operations:

[0075] The transfer probability is calculated based on the transfer comprehensive index corresponding to the adjacent cells;

[0076] Probabilistic selection is performed according to the transition probability corresponding to each adjacent cell among multiple adjacent cells to obtain a transition cell;

[0077] The update time corresponding to the transferred cell in multiple update times is used as the landing time.

[0078] Optionally, performing path parsing on multiple target path sequences to obtain path overlap and total time of multiple paths includes:

[0079] Confirm the number of paths for multiple target path sequences;

[0080] For each of the multiple target path sequences, the following operations are performed:

[0081] Extract the ath target coordinate group from the target path sequence, where the initial value of a is 1;

[0082] Extract the ath coordinate from the ath target coordinate group, where the ath coordinate is the target coordinate in the ath target coordinate group;

[0083] Summarizing the a-th coordinate to obtain multiple a-th coordinates, and extracting multiple reference horizontal coordinates, multiple reference vertical coordinates, and multiple reference vertical coordinates from the multiple a-th coordinates;

[0084] Calculating abscissa variance according to a plurality of reference abscissas, wherein the abscissa variance is the variance of the plurality of reference abscissas;

[0085] Obtaining a vertical coordinate variance and a vertical coordinate variance based on a plurality of reference vertical coordinates and a plurality of reference vertical coordinates respectively;

[0086] Calculate the average coordinate variance according to the horizontal coordinate variance, the vertical coordinate variance, and the vertical coordinate variance, wherein the average coordinate variance is the average of the horizontal coordinate variance, the vertical coordinate variance, and the vertical coordinate variance;

[0087] Let A = a + 1, with A as a, and return to the step of determining the number of paths in the plurality of target path sequences until A = M, where M is the number of paths;

[0088] Summarize the average coordinate variances to obtain multiple average coordinate variances, and calculate the path coincidence degree based on the multiple average coordinate variances;

[0089] The following operations are performed on each of the multiple target path sequences:

[0090] extracting a plurality of target times from a plurality of target coordinate groups in a target path sequence;

[0091] Calculate the total path time based on multiple target times, where the total path time is the sum of the multiple target times;

[0092] Sum up the total path times to get the total times of multiple paths.

[0093] Optionally, performing pheromone optimization on the negative marker space according to the total times of the multiple paths and the multiple target path sequences to obtain the path marker space includes:

[0094] Target path sequences are sequentially extracted from multiple target path sequences, and the following operations are performed on the extracted target path sequences:

[0095] Calculate the target pheromone increment based on the initial pheromone concentration and the total path time corresponding to the target path sequence in multiple path total times;

[0096] extracting a plurality of target coordinates from a plurality of target coordinate groups in a target path sequence;

[0097] For each of the multiple target coordinates, perform the following operations:

[0098] Identify the target cell in the negative marker space based on the target coordinates;

[0099] Identify the target pheromone concentration of the target cell, and calculate the replacement pheromone concentration based on the target pheromone concentration and the target pheromone increment;

[0100] Based on the replacement pheromone concentration, the target cell is updated with pheromone to obtain a supplemented cell;

[0101] Summarize the supplementary cells to obtain multiple supplementary cells;

[0102] The negative mark space is updated based on the multiple supplementary cells to obtain a supplementary three-dimensional space;

[0103] The supplemented three-dimensional space is used as a negative labeling space, and the step of sequentially extracting target path sequences from multiple target path sequences is returned to until all multiple target path sequences are extracted, and the supplemented three-dimensional space is used as a path labeling space.

[0104] To achieve the above objectives, the present invention further provides a multi-UAV navigation path planning system based on multi-task parameters, comprising:

[0105] a mapping space construction module, configured to obtain a three-dimensional mapping space, wherein the three-dimensional mapping space is in the shape of a cube and includes: a spatial rectangular coordinate system and a plurality of cells, wherein the cell is in the shape of a cube, a unit length of the spatial rectangular coordinate system is a side length of the cell, each of the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstruction cells;

[0106] An environmental data labeling module is used to identify the current wind speed, current gust vector, current rainfall, and the UAV mission sequence, wherein the UAV mission sequence includes: K mission coordinate groups, each mission coordinate group includes: starting coordinates and ending coordinates; obtain multiple historical fault groups, wherein the historical fault groups include: historical fault coordinates and historical fault counts; based on a preset initial pheromone concentration and the multiple historical fault groups, negatively label the three-dimensional mapping space with pheromones to obtain a negatively labeled space;

[0107] The ant colony algorithm optimization module is used to extract the j-th task coordinate group from the UAV task sequence, where the initial value of j is 1, and perform an ant colony simulation on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-built optimization heuristic function, the pre-built path coordinate group sequence set and the negative label space to obtain an updated coordinate group sequence set, where the updated coordinate group sequence set includes: multiple optimal path sequences;

[0108] The optimal path generation module is used to use the updated coordinate group sequence set as the path coordinate group sequence set, let J=j+1, take J as j, return to the step of extracting the j-th task coordinate group from the UAV task sequence, until j=K, use the updated coordinate group sequence set to generate multiple optimal paths, wherein the optimal paths correspond to the task coordinate groups one-to-one, and complete the multi-UAV navigation path planning based on the multiple optimal paths.

[0109] In order to solve the above problem, the present invention further provides an electronic device, comprising:

[0110] a memory storing at least one instruction;

[0111] The processor executes the instructions stored in the memory to implement the multi-UAV navigation path planning method based on multi-task parameters described above.

[0112] In order to solve the above problems, the present invention also provides a computer-readable storage medium, which stores at least one instruction, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned multi-UAV navigation path planning method based on multi-task parameters.

[0113] In order to solve the problems described in the background technology, the present invention obtains a three-dimensional mapping space, wherein the shape of the three-dimensional mapping space is a cube, and the three-dimensional mapping space includes: a spatial rectangular coordinate system and a plurality of cells, wherein the shape of the cell is a cube, the unit length of the spatial rectangular coordinate system is the side length of the cell, each of the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstacle cells. It can be seen that the embodiment of the present invention maps the terrain features and spatial structure of the target mission area in reality into a three-dimensional mapping space, thereby converting the real spatial environment information into a mathematical model that can be processed by a computer, and providing a computable space for subsequent path planning. The current wind speed, current gust vector, current rainfall and UAV task sequence are determined based on the time structure, wherein the UAV task sequence includes: K task coordinate groups, and the task coordinate group includes: starting coordinates and end point coordinates. It can be seen that the embodiment of the present invention provides real-time environmental parameters for subsequent path planning by obtaining the current wind speed, current gust vector and current rainfall, thereby enhancing the reliability and safety of path planning. At the same time, by obtaining the UAV task sequence, it is convenient to perform path planning in sequence according to the priority of the tasks in the UAV task sequence, and obtain multiple historical fault groups, wherein the historical fault group includes: historical fault coordinates and historical fault times, and performs negative pheromone on the three-dimensional mapping space based on the preset initial pheromone concentration and multiple historical fault groups. Marking, obtaining a negative marking space, it can be seen that the embodiment of the present invention introduces historical fault data and uses historical fault data to mark the three-dimensional mapping space with negative pheromones, so that the ant colony algorithm execution process automatically avoids high-risk areas, improves the reliability and safety of path planning, and extracts the j-th task coordinate group from the UAV task sequence, wherein the initial value of j is 1, and the j-th task coordinate group is optimized based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative marking space. Ant colony simulation is performed to obtain an updated coordinate group sequence set, wherein the updated coordinate group sequence set includes: multiple optimal path sequences. It can be seen that the embodiment of the present invention uses the improved ant colony algorithm to consider the environment The optimal path is searched on the basis of the factors and the mutual interference factors of multiple UAVs, thereby improving the accuracy and reliability of path planning. The updated coordinate group sequence set is used as the path coordinate group sequence set, and J=j+1 is set. J is used as j, and the step of extracting the jth task coordinate group from the UAV task sequence is returned until j=K. The updated coordinate group sequence set is used to generate multiple optimal paths, wherein the optimal paths correspond to the task coordinate groups one by one, and the navigation path planning of multiple UAVs is completed based on the multiple optimal paths. It can be seen that the embodiment of the present invention extracts the task coordinate groups from the UAV task sequence in sequence through a cyclic process and generates the optimal paths one by one, thereby avoiding the overlap or interference of the paths of multiple UAVs and improving the safety of multiple UAVs when performing tasks at the same time.Therefore, the present invention can improve the accuracy and reliability of path planning and enhance the safety of multiple drones performing tasks simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0114] Figure 1 A schematic diagram of a flow chart of a multi-UAV navigation path planning method based on multi-task parameters provided by one embodiment of the present invention;

[0115] Figure 2 A functional module diagram of a multi-UAV navigation path planning system based on multi-task parameters provided by one embodiment of the present invention;

[0116] Figure 3 A schematic structural diagram of an electronic device for implementing the multi-UAV navigation path planning method based on multi-task parameters provided by one embodiment of the present invention.

[0117] Description of reference numerals:

[0118] 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0119] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0120] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0121] The embodiments of the present application provide a multi-UAV navigation path planning method based on multi-task parameters. The execution subject of the multi-UAV navigation path planning method based on multi-task parameters includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiments of the present application. In other words, the multi-UAV navigation path planning method based on multi-task parameters can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0122] Reference Figure 1 FIG. 1 is a flow chart of a multi-UAV navigation path planning method based on multi-task parameters according to an embodiment of the present invention. In this embodiment, the multi-UAV navigation path planning method based on multi-task parameters includes:

[0123] S1. Obtain a three-dimensional mapping space, wherein the shape of the three-dimensional mapping space is a cube, and the three-dimensional mapping space includes: a spatial rectangular coordinate system and a plurality of cells, wherein the shape of the cell is a cube, the unit length of the spatial rectangular coordinate system is the side length of the cell, each cell in the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstacle cells.

[0124] It should be explained that the three-dimensional mapping space is a three-dimensional model used to reflect the topographic features and spatial structure of the target mission area. By mapping the target mission area into the three-dimensional mapping space, real-world spatial environment information is converted into a computer-processable mathematical model. Furthermore, the three-dimensional mapping space is generated by the UAV control center staff who pre-scan the target mission area using a LiDAR point cloud scanning system to obtain three-dimensional point cloud data for the target mission area. This data is then input into a computer's three-dimensional mapping software (such as OctoMap, RTAB-Map, and Point Cloud Library) for modeling and processing, ultimately generating the three-dimensional mapping space. The target mission area is the maximum spatial area within which multiple UAVs are allowed to move when performing a mission in reality, and the target mission area is shaped like a cube in reality. The UAV control center is responsible for controlling multiple UAVs. The techniques for scanning the target mission area using a LiDAR point cloud scanning system to obtain three-dimensional point cloud data for the target mission area, and then inputting this data into a computer's three-dimensional mapping software (such as OctoMap, RTAB-Map, and Point Cloud Library) for modeling and processing, ultimately generating the three-dimensional mapping space, are prior art and will not be further described in detail in the present embodiments.

[0125] It is understandable that in order to reduce the amount of data and computational complexity in the three-dimensional mapping space modeling process, when using the target task area to construct the three-dimensional mapping space, the resolution conversion rate of mapping the target task area to the three-dimensional mapping space is pre-set. For example, when the resolution conversion rate is 100, the corresponding distance between two points in the target task area that are 1 meter apart in the three-dimensional mapping space is 1 centimeter, that is, every 1 centimeter mapping unit corresponds to a distance of 1 meter in the real space.

[0126] It should be understood that when constructing the 3D mapping space, it is divided into multiple cubes of equal volume according to the principle of uniform division and the preset cell side lengths. These multiple cubes of equal volume are the multiple cells. For example, if the preset cell side length is 1 cm and the total volume of the 3D mapping space is 1 cubic meter, then the 3D mapping space includes a total of 1,000,000 cells.

[0127] It can be understood that a blank cell refers to a cell in the three-dimensional mapping space that is not occupied by any obstacles, representing a space where the drone can pass freely, and an obstacle cell refers to a cell where an obstacle exists, representing a space where the drone cannot pass.

[0128] It should be understood that any vertex of the three-dimensional mapping space is taken as the origin, and the three adjacent sides of the vertex are used as the horizontal axis, vertical axis and vertical axis respectively to construct the spatial rectangular coordinate system. The unit length of the spatial rectangular coordinate system is the side length of the cell, and the coordinates corresponding to the geometric center of each cell in the spatial rectangular coordinate system are used to indicate the position of the cell. Therefore, each cell in the three-dimensional mapping space has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system. For example: the coordinates of the cell closest to the origin are (0.5, 0.5, 0.5), and the coordinates of the cell adjacent to the cell closest to the origin and directly above the cell closest to the origin are (0.5, 0.5, 1.5).

[0129] S2. Confirm the current wind speed, current gust vector, current rainfall and UAV mission sequence, wherein the UAV mission sequence includes: K mission coordinate groups, and the mission coordinate group includes: starting coordinates and end coordinates.

[0130] It should be explained that the current wind speed refers to the wind speed in the target mission area at this time. Optionally, multiple wind speeds in the target mission area at this time are collected by multiple wind speed sensors, and the average of the multiple wind speeds is calculated to obtain the current wind speed. The current rainfall refers to the precipitation in the target mission area at this time. The current gust vector is a unit vector in the three-dimensional mapping space that reflects the wind direction in the target mission area at this time. For example, if the wind direction in the target mission area at this time is north wind, and the positive direction of the horizontal axis in the three-dimensional mapping space corresponds to the direction from south to north in reality, then the current gust vector is (-1, 0, 0). If the wind direction in the target mission area at this time is south wind, then the current gust vector is (1, 0, 0). Optionally, the wind direction and rainfall are obtained by querying the weather in the target mission area on the website of the China Meteorological Administration.

[0131] For example, if there are multiple drones about to perform tasks in the target mission area, the position of one of the drones in the target mission area at this time, the coordinates of the corresponding cell in the three-dimensional mapping space are the starting coordinates, and the drone needs to reach the position in the target mission area when performing the task, and the coordinates of the corresponding cell in the three-dimensional mapping space are the end coordinates. The starting coordinates and the end coordinates constitute the task coordinate group, and one drone corresponds to one task coordinate group. If there are K drones, there are K task coordinate groups in total. The K task coordinate groups are sorted according to the priority of executing the task to obtain a drone task sequence, and the priority of executing the task is determined by the staff of the drone control center.

[0132] S3. Acquire multiple historical fault groups, where the historical fault groups include historical fault coordinates and historical fault times.

[0133] It can be understood that the coordinates of the cells corresponding to the locations where the drone has historically failed in the target mission area in the three-dimensional mapping space are the historical failure coordinates. The number of historical failures refers to the number of times the failure has occurred at the historical failure coordinates. For example, the number of failures at the location corresponding to the coordinates (12, 3, 9) is 1, then the historical failure group is {(12, 3, 9), 1}, the number of failures at the location corresponding to the coordinates (32, 1, 12) is 2, then the historical failure group is {(32, 1, 12), 2}. The records of multiple drone failures in history are counted to generate multiple historical failure groups.

[0134] S4. Based on the preset initial pheromone concentration and multiple historical fault groups, negative pheromone marking is performed on the three-dimensional mapping space to obtain a negative marking space.

[0135] Specifically, the negative pheromone marking of the three-dimensional mapping space based on the preset initial pheromone concentration and multiple historical fault groups to obtain the negative marking space includes:

[0136] Perform the following operations on each of the multiple historical fault groups:

[0137] Identify the fault cells in the three-dimensional mapping space based on the historical fault coordinates in the historical fault group;

[0138] Identifying a plurality of dangerous cells in a three-dimensional mapping space based on the faulty cell, wherein the dangerous cells are adjacent to the faulty cell in the three-dimensional mapping space;

[0139] The danger pheromone concentration is calculated based on the initial pheromone concentration and the number of historical failures in the historical failure group. The calculation formula is as follows:

[0140]

[0141] in, is the concentration of danger pheromone, is the number of historical failures, is the initial pheromone concentration, is a natural constant;

[0142] Calculate the barrier pheromone concentration based on the danger pheromone concentration, where the barrier pheromone concentration is half of the danger pheromone concentration;

[0143] Marking a fault cell based on the obstacle pheromone concentration to obtain an obstacle-marked cell, and marking each of a plurality of dangerous cells based on the danger pheromone concentration to obtain a plurality of danger-marked cells;

[0144] Aggregating the obstacle marking cells and the plurality of danger marking cells to obtain a plurality of exclusion cells, wherein the exclusion cells are obstacle marking cells or danger marking cells;

[0145] Eliminate multiple exclusion cells from multiple cells in the three-dimensional mapping space to obtain multiple basic cells;

[0146] Each of the plurality of basic cells is marked based on the initial pheromone concentration to obtain a plurality of initially marked cells;

[0147] Aggregating multiple exclusion cells and multiple initial mark cells to obtain multiple initial cells, wherein the initial cells are exclusion cells or initial mark cells;

[0148] A negative label space is identified based on multiple initial cells and a three-dimensional mapping space.

[0149] It should be explained that identifying the fault cell in the three-dimensional mapping space based on the historical fault coordinates in the historical fault group means: taking the cell with the historical fault coordinates among the multiple cells in the three-dimensional mapping space as the fault cell.

[0150] It is understood that marking the faulty cell based on the obstacle pheromone concentration to obtain the obstacle-marked cell means: using the obstacle pheromone concentration value to identify the faulty cell, and the obstacle-marked cell is the marked faulty cell. Marking each of the multiple dangerous cells based on the danger pheromone concentration to obtain the multiple danger-marked cells means: using the danger pheromone concentration value to identify each of the multiple dangerous cells, and the multiple danger-marked cells are the marked multiple dangerous cells.

[0151] It should be understood that the method for marking each of the multiple basic cells based on the initial pheromone concentration to obtain multiple initially marked cells is the same as the method for marking each of the multiple dangerous cells based on the danger pheromone concentration to obtain multiple dangerous marked cells, and will not be repeated here. The initial pheromone concentration is manually set by the staff of the drone control center, and optionally, the initial pheromone concentration is 1.

[0152] It is understandable that the identification of the negatively labeled space based on the multiple initial cells and the three-dimensional mapping space means that after it is confirmed that the multiple cells in the three-dimensional mapping space are converted into the multiple initial cells, the three-dimensional mapping space at this time is the negatively labeled space.

[0153] S5. Extract the j-th task coordinate group from the UAV task sequence, where the initial value of j is 1, and perform an optimized ant colony simulation on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative label space to obtain an updated coordinate group sequence set, where the updated coordinate group sequence set includes: multiple optimal path sequences.

[0154] It should be understood that the embodiment of the present invention improves the ant colony algorithm while taking into account the situation where multiple drones are running at the same time. The relationship between the positions and times of multiple drones that have completed path planning is recorded through a path coordinate group sequence set, thereby preventing subsequent drones from overlapping with the paths of other drones when performing path planning, and preventing multiple drones from interfering with each other when performing tasks. At the same time, the ant colony algorithm is further improved by introducing an optimization heuristic function, taking into account the impact of real-time environmental factors on path planning, thereby improving the accuracy and reliability of path planning for multiple drones using the ant colony algorithm.

[0155] It should be explained that the pre-constructed path coordinate group sequence set is an empty set, and its main function is to save the updated memory group sequences of multiple drones that have completed path planning. For example: when the first task coordinate group is extracted, the path coordinate group sequence set is an empty set, but the path coordinate group sequence set is subsequently updated according to the optimal path sequence to obtain an updated coordinate group sequence set, and the updated coordinate group sequence set is used as the path coordinate group sequence set, and the step of extracting the jth task coordinate group from the drone task sequence is returned, that is, when the second task coordinate group is extracted again, the path coordinate group sequence set is no longer an empty set, and the updated memory group sequence corresponding to the first task coordinate group is saved. At this time, interference feature analysis can be performed according to the optimal path sequence in the path coordinate group sequence set, thereby preventing the path corresponding to the first task coordinate group from overlapping when the second task coordinate group is planned.

[0156] Specifically, the optimized ant colony simulation is performed on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set, and the negative label space to obtain the updated coordinate group sequence set, including:

[0157] According to the starting coordinates in the j-th task coordinate group, the starting cell is identified in the negative marking space, and the ending coordinates in the j-th task coordinate group are marked as the ending coordinates;

[0158] Generate multiple virtual ants, and perform the following operations on each of the multiple virtual ants:

[0159] Place the virtual ant into the starting cell to obtain the starting cell;

[0160] Identify multiple neighboring cells of the starting cell in the negatively marked space;

[0161] For each of multiple neighboring cells, perform the following operations:

[0162] Perform transfer strategy analysis on neighboring cells and starting cells based on current wind speed, current gust vector, current rainfall, optimization heuristic function, termination coordinates, preset number of loops, pre-built memory coordinate group sequence, preset start time and path coordinate group sequence set to obtain transfer comprehensive index and update time;

[0163] Summarize the transfer comprehensive indexes and update times respectively to obtain multiple transfer comprehensive indexes and multiple update times, wherein the transfer comprehensive indexes and update times are in one-to-one correspondence with adjacent cells;

[0164] Based on multiple transfer comprehensive indexes and multiple update times, multiple adjacent cells are probabilistically transferred to obtain the transferred cells and landing time;

[0165] Move the virtual ant to the transfer cell, obtain the landing cell, and identify the landing coordinates of the landing cell in the negative mark space;

[0166] Combine the landing point coordinates and the landing time to obtain a landing point coordinate group;

[0167] Using the landing point coordinate group to update the memory coordinate group sequence, to obtain an updated memory group sequence;

[0168] Determine whether the landing point coordinates are equal to the ending coordinates;

[0169] If the landing point coordinates are not equal to the ending coordinates, the landing point cell is used as the starting cell, the updated memory group sequence is used as the memory coordinate group sequence, the landing time is used as the starting time, and the process returns to the step of identifying multiple neighboring cells of the starting cell in the negative mark space until the landing point coordinates are equal to the ending coordinates;

[0170] If the landing point coordinates are equal to the ending coordinates, the updated memory group sequence is used as the target path sequence;

[0171] Summarizing the target path sequences to obtain multiple target path sequences, wherein the target path sequences include: multiple target coordinate groups, and the target coordinate groups include: target coordinates and target time;

[0172] Perform path analysis on multiple target path sequences to obtain the path overlap and the total time of multiple paths;

[0173] Determine the path overlap with a preset path threshold and compare the number of cycles with a preset cycle threshold;

[0174] If the path overlap is less than the path threshold and the number of cycles is less than the cycle threshold, pheromone optimization is performed on the negative marking space according to the total time of multiple paths and the multiple target path sequences to obtain a path marking space, the number of updates is calculated according to the number of cycles, the path marking space is used as the negative marking space, the number of updates is used as the number of cycles, and the step of generating multiple virtual ants is returned to until the path overlap is greater than or equal to the path threshold or the number of cycles is greater than or equal to the cycle threshold;

[0175] If the path overlap is greater than or equal to the path threshold or the number of cycles is greater than or equal to the cycle threshold, the optimal path sequence is determined based on the total time of multiple paths and multiple target path sequences, and the path coordinate group sequence set is updated according to the optimal path sequence to obtain an updated coordinate group sequence set.

[0176] It should be explained that the method of identifying the starting cell in the negative marking space based on the starting coordinates in the j-th task coordinate group is the same as the method of identifying the fault cell in the three-dimensional mapping space based on the historical fault coordinates in the historical fault group, and will not be repeated here. Neighboring cells refer to: initial cells in the negative marking space that are adjacent to the faces, edges and points of the starting cell among multiple initial cells. For example: each cell in the 3×3 cell area centered on the starting cell is adjacent to the faces, edges and points of the starting cell, that is, one starting cell corresponds to a total of 26 neighboring cells.

[0177] It is understandable that the multiple virtual ants are path search individuals generated based on the ant colony algorithm, which are used to simulate path selection behavior in the negative marking space. By simulating the parallel search process of multiple virtual ants, multiple feasible paths from the starting coordinates to the ending coordinates in the negative marking space are explored.

[0178] It should be explained that the landing point coordinates are the coordinates corresponding to the position of the landing point cell in the negative mark space in the spatial rectangular coordinate system.

[0179] It should be understood that the memory coordinate group sequence includes: multiple memory coordinate groups, the number of loops, the starting time and the memory coordinate group sequence will be continuously updated in a loop when the step of returning to the multiple neighboring cells of the starting cell identified in the negative mark space is executed. During the first loop, the initial values ​​of the number of loops, the starting time and the memory coordinate group sequence are as follows: the preset number of loops is 1, the preset starting time is 0 seconds, and the pre-constructed memory coordinate group sequence is: {[(0,0,0),0], [(0.5,0.5,0.5),0]}.

[0180] For example, if the landing point coordinates are (1.5, 1.5, 1.5) and the landing time is 2 seconds, the landing point coordinate group is: [(1.5, 1.5, 1.5), 2]. The use of the landing point coordinate group to update the memory coordinate group sequence means: placing the landing point coordinate group as a memory coordinate group in the memory coordinate group sequence, and the sorting position of the landing point coordinate group in the memory coordinate group sequence is the last. In the first cycle, since the memory coordinate group sequence is: {[(0, 0, 0), 0], [(0.5, 0.5, 0.5), 0]}, the updated memory group sequence is: {[(0, 0, 0), 0], [(0.5, 0.5, 0.5), 0], [(1.5, 1.5, 1.5), 2]}. If the ending coordinates are (12.5, 9.5, 17.5), the landing point coordinates are not equal to the ending coordinates, and the landing unit is updated. The grid is used as the starting cell, the updated memory group sequence is used as the memory coordinate group sequence, the landing time is used as the starting time, and the step of identifying multiple neighboring cells of the starting cell in the negative mark space is returned until the landing coordinates are equal to the ending coordinates. At this time, after multiple iterative updates, the updated memory group sequence is: {[(0,0,0),0], [(0.5,0.5,0.5),0], [(1.5,1.5,1.5),2]……[(12.5,9.5,17.5),19]}, and the updated memory group sequence at this time is used as the target path sequence, and one virtual ant corresponds to one target path sequence, so multiple target path sequences are obtained by summarizing.

[0181] It is understood that calculating the update count based on the number of loops means adding one to the loop count to obtain the update count. For example, if the loop count is 1, the update count is 2. Determining the optimal path sequence based on multiple path total times and multiple target path sequences means determining the target path sequence corresponding to the shortest path total time among the multiple target path sequences as the optimal path sequence. Updating the path coordinate group sequence set based on the optimal path sequence means placing the optimal path sequence as the path coordinate group sequence in the path coordinate group sequence set. Preferably, the path threshold is 0.5, and the loop threshold is 500.

[0182] Specifically, the transfer strategy analysis is performed on the neighboring cells and the starting cell based on the current wind speed, the current gust vector, the current rainfall, the optimization heuristic function, the end coordinates, the preset number of cycles, the pre-built memory coordinate group sequence, the preset start time and the path coordinate group sequence set to obtain the transfer comprehensive index and update time, including:

[0183] Identify the neighboring coordinates and starting coordinates of the neighboring cells and starting cell in the negative label space respectively;

[0184] The far distance is calculated based on the adjacent coordinates and the end coordinates. The calculation formula is as follows:

[0185]

[0186] in, For the distance, 、 and are the horizontal, vertical and vertical coordinates of the terminal coordinates respectively. 、 and are the horizontal, vertical and vertical coordinates in the adjacent coordinates respectively;

[0187] Get the nearest distance based on the neighboring coordinates and the starting coordinates;

[0188] The moving time is calculated based on the preset resolution conversion rate, the nearest distance, and the preset average speed. The calculation formula is as follows:

[0189]

[0190] in, is the moving time, For close distance, To resolve the conversion rate, is the average speed;

[0191] Calculate the update time based on the moving time and the starting time, where the update time is the sum of the moving time and the starting time;

[0192] Determine whether the adjacent cell is the obstacle cell, and if the adjacent cell is the obstacle cell, use the preset zero probability as the transfer comprehensive index;

[0193] Otherwise, a motion vector is determined based on the neighboring coordinates and the starting coordinates, where the motion vector is as follows:

[0194]

[0195] in, is the movement vector, 、 and They are the horizontal coordinate, vertical coordinate and vertical coordinate of the starting point coordinate respectively;

[0196] Identifying a last coordinate group in the memory coordinate group sequence, wherein the last coordinate group is the last memory coordinate group in the memory coordinate group sequence;

[0197] Identifying an advance coordinate group in the memory coordinate group sequence based on the last coordinate group, wherein the last coordinate group is adjacent to the advance coordinate group in the memory coordinate group sequence and lags behind the advance coordinate group;

[0198] Extract the advance coordinates from the advance coordinate group, extract the final coordinates from the final coordinate group, obtain the memory vector based on the advance coordinates and the final coordinates, and calculate the corner energy consumption index based on the memory vector and the movement vector. The calculation formula is as follows:

[0199]

[0200] in, is the corner energy consumption index, is the memory vector, is the modulus of the motion vector, is the modulus of the memory vector, is the inverse cosine function, is the preset unit energy consumption value;

[0201] The environmental impact index is calculated based on the movement vector, current wind speed, current gust vector, and current rainfall. The calculation formula is as follows:

[0202]

[0203] in, is the environmental impact index, and are the current rainfall and current wind speed, respectively. and are the current gust vector and the preset rise vector respectively, is the movement vector, is the modulus of the motion vector, and are the preset lift resistance coefficient and the preset wind speed resistance coefficient, and are the modulus of the lift vector and the modulus of the current gust vector respectively;

[0204] Identify neighboring pheromone concentrations of neighboring cells;

[0205] Based on the update time and adjacent coordinates, the interference characteristics of the path coordinate group sequence set are analyzed to obtain the mutual interference index;

[0206] The number of cycles, neighboring pheromone concentration, near distance, far distance, mutual interference index, corner energy consumption index and environmental impact index are substituted into the optimization heuristic function to obtain the transfer comprehensive index.

[0207] It should be explained that the adjacent coordinates are the coordinates of the adjacent cells in the negative marker space in the spatial rectangular coordinate system. The starting coordinates are the coordinates of the starting cell in the negative marker space in the spatial rectangular coordinate system.

[0208] In an embodiment of the present invention, since the spatial rectangular coordinate system includes: an x-axis, a y-axis, and a z-axis, the horizontal axis, vertical axis, and vertical axis correspond to the x-axis, y-axis, and z-axis, respectively, and the horizontal coordinate, vertical coordinate, and vertical coordinate correspond to the coordinate on the x-axis, the coordinate on the y-axis, and the coordinate on the z-axis in the spatial rectangular coordinate system, respectively.

[0209] It will be appreciated that the method for obtaining the near-field distance based on the neighboring coordinates and the starting coordinates is the same as the method for calculating the far-field distance based on the neighboring coordinates and the ending coordinates, and will not be further described here. The preset average speed refers to the average speed of the drone in straight-line flight at rated power. The preset unit energy consumption value refers to the average energy consumed per degree of change in flight direction at rated power. The preset lift drag coefficient refers to the energy consumed by the drone in a straight-line ascent of 1 meter at rated power. The preset wind speed drag coefficient refers to the energy consumed by the drone in a straight-line flight of 1 meter against the wind at rated power and a wind speed of 1 m / s. The average speed, unit energy consumption value, lift drag coefficient, and wind speed drag coefficient are all measured by drone control center staff through test flights of the drone. The preset zero return probability is 0%. The preset ascent vector is (0, 0, 1).

[0210] It should be explained that the advance coordinates refer to the coordinates in the advance coordinate group, and the last coordinates refer to the coordinates in the last coordinate group. For example, if the advance coordinate group is [(0.5, 0.5, 0.5), 0], the advance coordinates are (0.5, 0.5, 0.5); if the last coordinate group is [(1.5, 1.5, 1.5), 2], the last coordinates are (1.5, 1.5, 1.5).

[0211] It is understandable that the method of obtaining the memory vector based on the advance coordinates and the final coordinates is the same as the method of confirming the movement vector based on the adjacent coordinates and the starting point coordinates, and will not be repeated here. The corner energy consumption index reflects the electrical energy consumed by the drone due to the change of direction when it moves in reality according to the moving path of the virtual ant in the negative marking space. The larger the corner energy consumption index, the more electrical energy is consumed by the drone due to the change of direction when it moves in reality according to the moving path of the virtual ant in the negative marking space. The environmental impact index reflects the electrical energy consumed by the resistance of wind and rainfall in the environment when the drone moves in reality according to the moving path of the virtual ant in the negative marking space. The smaller the environmental impact index, the less electrical energy is consumed by the drone due to the resistance of wind and rainfall in the environment when it moves in reality according to the moving path of the virtual ant in the negative marking space.

[0212] It should be understood that since multiple cells in the negatively labeled space have already been marked using pheromone concentration, identifying the neighboring pheromone concentrations of neighboring cells simply means using the pheromone concentrations corresponding to the neighboring cells as the neighboring pheromone concentrations. The transfer comprehensive index is the value output by the optimization heuristic function after substituting the number of cycles, neighboring pheromone concentrations, near distance, far distance, mutual interference index, corner energy consumption index, and environmental impact index into the optimization heuristic function.

[0213] Specifically, the interference feature analysis of the path coordinate group sequence set based on the update time and the adjacent coordinates to obtain the mutual interference index includes:

[0214] Determine whether the path coordinate group sequence set is an empty set. If the path coordinate group sequence set is an empty set, use the preset initial interference index as the mutual interference index. Otherwise, perform the following operations on each path coordinate group sequence in the path coordinate group sequence set:

[0215] identifying close coordinate groups in the path coordinate group sequence based on update time;

[0216] Extracting close coordinates and close time from the close coordinate group;

[0217] The mutual interference factor is calculated based on the update time, adjacent coordinates, close coordinates and close time. The calculation formula is as follows:

[0218]

[0219] in, is the mutual interference factor, and They are update time and close time respectively. 、 and are the horizontal, vertical and vertical coordinates in the close coordinates respectively. is the hyperbolic tangent function;

[0220] The mutual interference factors are aggregated to obtain multiple mutual interference factors, and a mutual interference index is calculated according to the multiple mutual interference factors, wherein the mutual interference index is the sum of the multiple mutual interference factors.

[0221] It should be explained that the initial interference index is 0.

[0222] Exemplarily, when the path coordinate group sequence set is not an empty set, it indicates that the optimal path sequence has been placed as a path coordinate group sequence in the path coordinate group sequence set during the loop process, and the optimal path sequence is from multiple target path sequences, and the target path sequence includes: multiple target coordinate groups, that is, identifying the close coordinate group in the path coordinate group sequence based on the update time means: using the target coordinate group corresponding to the target time closest to the update time among the multiple target coordinate groups in the path coordinate group sequence as the close coordinate group. The close coordinates are the target coordinates in the close coordinate group, and the close time is the target time in the close coordinate group. For example: the path coordinate group sequence is: {[(0,0,0),0], [(0.5,0.5,0.5),0], [(1.5,1.5,1.5),2], [(1.5,2.5,2.5),2.7]……[(12.5,9.5,17.5),19], if the update time is 2.1 seconds, then the approximate coordinate group is: [(1.5,1.5,1.5),2], the approximate coordinates are (1.5,1.5,1.5), and the approximate time is 2 seconds.

[0223] It should be understood that the mutual interference factor reflects the relative proximity between the drone and other drones when the drone moves in reality according to the moving path of the virtual ants in the negatively marked space. The larger the mutual interference factor, the greater the relative proximity between the drone and other drones when the drone moves in reality according to the moving path of the virtual ants in the negatively marked space, that is, the closer the distance between the drone and other drones. The mutual interference index is used to evaluate the interference risk caused by the overlap of the paths of drones with other drones during mission execution.

[0224] In detail, the optimization heuristic function is as follows:

[0225]

[0226] in, To optimize the heuristic function, is the corner energy consumption index, is the mutual interference index, is the neighboring pheromone concentration, is the number of cycles, is the cycle threshold, For the distance, For close distance.

[0227] It can be understood that the transfer comprehensive index reflects the priority of the virtual ant moving from the starting cell to the adjacent cell corresponding to the transfer comprehensive index. The larger the transfer comprehensive index, the higher the priority of the virtual ant moving from the starting cell to the adjacent cell corresponding to the transfer comprehensive index.

[0228] It should be understood that the embodiment of the present invention introduces the number of cycles into the optimization heuristic function. When the number of cycles is too small, the near distance, far distance, mutual interference index, corner energy consumption index and environmental impact index are used as the main influencing factors to prevent the multiple paths generated by multiple virtual ants from overlapping prematurely under the influence of pheromones and falling into local optimality. When the number of cycles is too large, the concentration of neighboring pheromones is used as the main influencing factor, thereby promoting the multiple paths generated by multiple virtual ants to converge as soon as possible, thereby improving the planning speed of the optimal path.

[0229] Specifically, the probabilistic transfer of multiple adjacent cells based on multiple transfer comprehensive indexes and multiple update times to obtain the transferred cells and landing times includes:

[0230] For each of multiple neighboring cells, perform the following operations:

[0231] The transition probability is calculated based on the transition comprehensive index corresponding to the adjacent cells. The calculation formula is as follows:

[0232]

[0233] in, is the transition probability, is the transfer comprehensive index corresponding to the adjacent cells, For multiple adjacent cells The transfer comprehensive index corresponding to the adjacent cells, is the number of adjacent cells in the plurality of adjacent cells;

[0234] Probabilistic selection is performed according to the transition probability corresponding to each adjacent cell among multiple adjacent cells to obtain a transition cell;

[0235] The update time corresponding to the transferred cell in multiple update times is used as the landing time.

[0236] It should be understood that the probabilistic selection based on the transition probability corresponding to each of the multiple neighboring cells to obtain the transfer cell means: using the roulette method to randomly select a neighboring cell from the multiple neighboring cells as the transfer cell according to the transition probability corresponding to each of the multiple neighboring cells. The roulette method is a random selection mechanism based on probability distribution commonly used in ant colony algorithms. By simulating the process of a turntable pointer falling on a certain area, the probability of each candidate being selected is proportional to its weight (here, the transition probability).

[0237] Specifically, performing path parsing on multiple target path sequences to obtain path overlap and the total time of multiple paths includes:

[0238] Confirm the number of paths for multiple target path sequences;

[0239] For each of the multiple target path sequences, the following operations are performed:

[0240] Extract the ath target coordinate group from the target path sequence, where the initial value of a is 1;

[0241] Extract the ath coordinate from the ath target coordinate group, where the ath coordinate is the target coordinate in the ath target coordinate group;

[0242] Summarizing the a-th coordinate to obtain multiple a-th coordinates, and extracting multiple reference horizontal coordinates, multiple reference vertical coordinates, and multiple reference vertical coordinates from the multiple a-th coordinates;

[0243] Calculating abscissa variance according to a plurality of reference abscissas, wherein the abscissa variance is the variance of the plurality of reference abscissas;

[0244] Obtaining a vertical coordinate variance and a vertical coordinate variance based on a plurality of reference vertical coordinates and a plurality of reference vertical coordinates respectively;

[0245] Calculate the average coordinate variance according to the horizontal coordinate variance, the vertical coordinate variance, and the vertical coordinate variance, wherein the average coordinate variance is the average of the horizontal coordinate variance, the vertical coordinate variance, and the vertical coordinate variance;

[0246] Let A = a + 1, with A as a, and return to the step of determining the number of paths in the plurality of target path sequences until A = M, where M is the number of paths;

[0247] Summarize the average coordinate variances to obtain multiple average coordinate variances, and calculate the path coincidence based on the multiple average coordinate variances. The calculation formula is as follows:

[0248]

[0249] in, is the path overlap, is the first of multiple average coordinate variances mean coordinate variances, m is the number of mean coordinate variances in the multiple mean coordinate variances;

[0250] The following operations are performed on each of the multiple target path sequences:

[0251] extracting a plurality of target times from a plurality of target coordinate groups in a target path sequence;

[0252] Calculate the total path time based on multiple target times, where the total path time is the sum of the multiple target times;

[0253] Sum up the total path times to get the total times of multiple paths.

[0254] It should be explained that the number of paths is the number of target path sequences in the plurality of target path sequences. The reference abscissa, reference ordinate, and reference ordinate are the abscissa, ordinate, and ordinate in the a-th coordinate, respectively.

[0255] It is understandable that the method of obtaining the vertical coordinate variance based on multiple reference vertical coordinates and the method of obtaining the vertical coordinate variance based on multiple reference vertical coordinates are both the same as the method of calculating the horizontal coordinate variance based on multiple reference horizontal coordinates, and are not described in detail here.

[0256] In detail, the pheromone optimization of the negative marking space according to the total time of multiple paths and the multiple target path sequences to obtain the path marking space includes:

[0257] Target path sequences are sequentially extracted from multiple target path sequences, and the following operations are performed on the extracted target path sequences:

[0258] The target pheromone increment is calculated based on the initial pheromone concentration and the total path time corresponding to the target path sequence in multiple path total times. The calculation formula is as follows:

[0259]

[0260] in, is the target pheromone increment, is the total path time, is the preset reference total time;

[0261] extracting a plurality of target coordinates from a plurality of target coordinate groups in a target path sequence;

[0262] For each of the multiple target coordinates, perform the following operations:

[0263] Identify the target cell in the negative marker space based on the target coordinates;

[0264] Identify the target pheromone concentration of the target cell and calculate the replacement pheromone concentration based on the target pheromone concentration and the target pheromone increment. The calculation formula is as follows:

[0265]

[0266] in, To change the pheromone concentration, is the target pheromone concentration, is the preset replacement factor;

[0267] Based on the replacement pheromone concentration, the target cell is updated with pheromone to obtain a supplemented cell;

[0268] Summarize the supplementary cells to obtain multiple supplementary cells;

[0269] The negative mark space is updated based on the multiple supplementary cells to obtain a supplementary three-dimensional space;

[0270] The supplemented three-dimensional space is used as a negative labeling space, and the step of sequentially extracting target path sequences from multiple target path sequences is returned to until all multiple target path sequences are extracted, and the supplemented three-dimensional space is used as a path labeling space.

[0271] It should be explained that the reference total time is set by the staff of the drone control center based on the historical operation records of the drone. Optionally, the average of the total time of all paths recorded historically by the drone is used as the reference total time.

[0272] It should be understood that the extraction of multiple target coordinates from multiple target coordinate groups in the target path sequence means: extracting the target coordinates in each target coordinate group in the multiple target coordinate groups and summing them up to obtain multiple target coordinates. The identification of the target cell in the negative marker space based on the target coordinates means: using the initial cell with the target coordinates in the negative marker space as the target cell. The method for identifying the target pheromone concentration of the target cell is the same as the method for identifying the neighboring pheromone concentration of the neighboring cells, and will not be repeated here. The replacement coefficient is a value manually set by the staff of the drone control center. Preferably, the replacement coefficient is 0.2.

[0273] It is understood that since the target cell has already been identified by the target pheromone concentration, updating the target cell's pheromone based on the replacement pheromone concentration means erasing the target cell's original target pheromone concentration and re-identifying the target cell using the replacement pheromone concentration. Updating the negatively marked space based on multiple supplementary cells to obtain a supplementary three-dimensional space means sequentially extracting supplementary cells from the multiple supplementary cells and replacing the initial cells in the negatively marked space with the same coordinates as the supplementary cells with the extracted supplementary cells until all the supplementary cells have been extracted. The negatively marked space after being replaced by the multiple supplementary cells serves as the supplementary three-dimensional space.

[0274] S6. Use the updated coordinate group sequence set as the path coordinate group sequence set, let J = j + 1, and use J as j, return to the step of extracting the j-th task coordinate group from the UAV task sequence until j = K, and use the updated coordinate group sequence set to generate multiple optimal paths, where the optimal paths correspond one-to-one to the task coordinate groups.

[0275] In detail, the method of generating multiple optimal paths by using the updated coordinate group sequence set includes:

[0276] The following operations are performed for each optimal path sequence in the updated coordinate group sequence set:

[0277] Obtain target coordinate sequence using the optimal path sequence;

[0278] Generate the optimal path using the target coordinate sequence;

[0279] Summarize the optimal paths and obtain multiple optimal paths.

[0280] It should be explained that the method of obtaining a target coordinate sequence using an optimal path sequence refers to extracting multiple target coordinates from multiple target coordinate groups in the optimal path sequence, and sorting the multiple target coordinates according to the order of the multiple target coordinate groups in the optimal path sequence to obtain a target coordinate sequence. The method of generating an optimal path using a target coordinate sequence refers to fitting the multiple target coordinates in the target coordinate sequence into a continuous and smooth three-dimensional path curve in sequence using a cubic spline interpolation method, and the three-dimensional path curve is the optimal path. The technology of fitting the multiple target coordinates in the target coordinate sequence into a continuous and smooth three-dimensional path curve in sequence using a cubic spline interpolation method is prior art and will not be described in detail here.

[0281] For example, when j=K, that is, all K task coordinate groups have completed path planning, then the updated coordinate group sequence set can be used to generate multiple optimal paths, so that multiple drones can move according to the corresponding paths in reality according to their respective optimal paths.

[0282] S7. Complete multi-UAV navigation path planning based on multiple optimal paths.

[0283] In order to solve the problems described in the background technology, the present invention obtains a three-dimensional mapping space, wherein the shape of the three-dimensional mapping space is a cube, and the three-dimensional mapping space includes: a spatial rectangular coordinate system and a plurality of cells, wherein the shape of the cell is a cube, the unit length of the spatial rectangular coordinate system is the side length of the cell, each of the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstacle cells. It can be seen that the embodiment of the present invention maps the terrain features and spatial structure of the target mission area in reality into a three-dimensional mapping space, thereby converting the real spatial environment information into a mathematical model that can be processed by a computer, and providing a computable space for subsequent path planning. The current wind speed, current gust vector, current rainfall and UAV task sequence are determined based on the time structure, wherein the UAV task sequence includes: K task coordinate groups, and the task coordinate group includes: starting coordinates and end point coordinates. It can be seen that the embodiment of the present invention provides real-time environmental parameters for subsequent path planning by obtaining the current wind speed, current gust vector and current rainfall, thereby enhancing the reliability and safety of path planning. At the same time, by obtaining the UAV task sequence, it is convenient to perform path planning in sequence according to the priority of the tasks in the UAV task sequence, and obtain multiple historical fault groups, wherein the historical fault group includes: historical fault coordinates and historical fault times, and performs negative pheromone on the three-dimensional mapping space based on the preset initial pheromone concentration and multiple historical fault groups. Marking, obtaining a negative marking space, it can be seen that the embodiment of the present invention introduces historical fault data and uses historical fault data to mark the three-dimensional mapping space with negative pheromones, so that the ant colony algorithm execution process automatically avoids high-risk areas, improves the reliability and safety of path planning, and extracts the j-th task coordinate group from the UAV task sequence, wherein the initial value of j is 1, and the j-th task coordinate group is optimized based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative marking space. Ant colony simulation is performed to obtain an updated coordinate group sequence set, wherein the updated coordinate group sequence set includes: multiple optimal path sequences. It can be seen that the embodiment of the present invention uses the improved ant colony algorithm to consider the environment The optimal path is searched on the basis of the factors and the mutual interference factors of multiple UAVs, thereby improving the accuracy and reliability of path planning. The updated coordinate group sequence set is used as the path coordinate group sequence set, and J=j+1 is set. J is used as j, and the step of extracting the jth task coordinate group from the UAV task sequence is returned until j=K. The updated coordinate group sequence set is used to generate multiple optimal paths, wherein the optimal paths correspond to the task coordinate groups one by one, and the navigation path planning of multiple UAVs is completed based on the multiple optimal paths. It can be seen that the embodiment of the present invention extracts the task coordinate groups from the UAV task sequence in sequence through a cyclic process and generates the optimal paths one by one, thereby avoiding the overlap or interference of the paths of multiple UAVs and improving the safety of multiple UAVs when performing tasks at the same time.Therefore, the present invention can improve the accuracy and reliability of path planning and enhance the safety of multiple drones performing tasks simultaneously.

[0284] like Figure 2 , which is a functional module diagram of a multi-UAV navigation path planning system based on multi-task parameters provided by one embodiment of the present invention.

[0285] The multi-task parameter-based multi-UAV navigation path planning system 100 described in the present invention can be installed in an electronic device 1. Depending on the functionality implemented, the multi-task parameter-based multi-UAV navigation path planning system 100 can include a mapping space construction module 101, an environmental data labeling module 102, an ant colony algorithm optimization module 103, and an optimal path generation module 104. A module, also referred to as a unit, refers to a series of computer program segments that can be executed by an electronic device processor and perform a fixed function, and is stored in the electronic device's memory.

[0286] The mapping space construction module 101 is used to obtain a three-dimensional mapping space, wherein the three-dimensional mapping space is in the shape of a cube and includes: a spatial rectangular coordinate system and a plurality of cells, wherein the cell is in the shape of a cube, the unit length of the spatial rectangular coordinate system is the side length of the cell, each of the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstacle cells;

[0287] The environmental data labeling module 102 is used to determine the current wind speed, the current gust vector, the current rainfall, and the UAV task sequence, wherein the UAV task sequence includes: K task coordinate groups, each task coordinate group includes: a starting coordinate and an ending coordinate; obtain multiple historical fault groups, wherein the historical fault groups include: historical fault coordinates and historical fault counts; and perform negative pheromone labeling on the three-dimensional mapping space based on a preset initial pheromone concentration and the multiple historical fault groups to obtain a negative labeling space;

[0288] The ant colony algorithm optimization module 103 is used to extract the j-th task coordinate group from the UAV task sequence, where the initial value of j is 1, and perform an optimized ant colony simulation on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative label space to obtain an updated coordinate group sequence set, where the updated coordinate group sequence set includes: multiple optimal path sequences;

[0289] The optimal path generation module 104 is configured to use the updated coordinate group sequence set as the path coordinate group sequence set, set J=j+1, and use J as j, return to the step of extracting the j-th task coordinate group from the UAV task sequence until j=K, and use the updated coordinate group sequence set to generate multiple optimal paths, wherein the optimal paths correspond one-to-one to the task coordinate groups, and the navigation path planning of multiple UAVs is completed based on the multiple optimal paths.

[0290] In detail, the modules in the multi-task parameter-based multi-UAV navigation path planning system 100 in the embodiment of the present invention are used in the same manner as above. Figure 1 The multi-UAV navigation path planning method based on multi-task parameters described in the present invention has the same technical means and can produce the same technical effects, so it will not be repeated here.

[0291] like Figure 3 , which is a structural diagram of an electronic device 1 for implementing a multi-UAV navigation path planning method based on multi-task parameters provided by an embodiment of the present invention.

[0292] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a multi-UAV navigation path planning method program based on multi-task parameters.

[0293] The memory 11 includes at least one type of readable storage medium, including flash memory, a removable hard drive, a multimedia card, a card-type memory (e.g., SD or DX memory), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a removable hard drive of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in removable hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used to store application software installed in the electronic device 1 and various data, such as the code of a multi-unmanned aerial vehicle navigation path planning method based on multi-mission parameters, and can also be used to temporarily store data that has been output or is about to be output.

[0294] In some embodiments, the processor 10 may be comprised of an integrated circuit, such as a single packaged integrated circuit or a plurality of packaged integrated circuits with the same or different functions, including a combination of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control core (control unit) of the electronic device, connecting the various components of the electronic device using various interfaces and circuits. It executes programs or modules stored in the memory 11 (e.g., a multi-unmanned aerial vehicle navigation path planning method based on multi-task parameters) and accesses data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0295] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0296] Figure 3 Only the electronic device with components is shown, and it can be understood by those skilled in the art that Figure 3 The structure shown does not constitute a limitation on the electronic device 1 , and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0297] For example, although not shown, the electronic device 1 may further include a power supply (e.g., a battery) to power various components. Preferably, the power supply may be logically connected to the at least one processor 10 via a power management device, thereby enabling functions such as charge management, discharge management, and power consumption management via the power management device. The power supply may further include any components such as one or more DC or AC power supplies, a recharging device, a power failure detection circuit, a power converter or inverter, and a power status indicator. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which are not further described here.

[0298] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0299] Optionally, the electronic device 1 may further include a user interface, which may be a display or an input unit (such as a keyboard). Optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display may also be appropriately referred to as a display screen or a display unit, and is used to display information processed by the electronic device 1 and to display a visual user interface.

[0300] The multi-UAV navigation path planning method program based on multi-task parameters stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve the following:

[0301] Acquire a three-dimensional mapping space, wherein the three-dimensional mapping space is in the shape of a cube and includes: a spatial rectangular coordinate system and a plurality of cells, wherein the cell is in the shape of a cube, a unit length of the spatial rectangular coordinate system is a side length of the cell, each cell in the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstruction cells;

[0302] Determine the current wind speed, current gust vector, current rainfall, and the UAV mission sequence, wherein the UAV mission sequence includes: K mission coordinate groups, and the mission coordinate group includes: starting coordinates and end coordinates;

[0303] Acquire multiple historical fault groups, where the historical fault groups include: historical fault coordinates and historical fault times;

[0304] Based on the preset initial pheromone concentration and multiple historical fault groups, the three-dimensional mapping space is negatively pheromone-marked to obtain a negatively marked space;

[0305] Extract the jth task coordinate group from the UAV task sequence, where the initial value of j is 1;

[0306] An ant colony simulation is performed on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative label space to obtain an updated coordinate group sequence set, wherein the updated coordinate group sequence set includes: multiple optimal path sequences;

[0307] Using the updated coordinate group sequence set as the path coordinate group sequence set, let J = j + 1, let J be j, return to the step of extracting the j-th task coordinate group from the UAV task sequence, until j = K, and use the updated coordinate group sequence set to generate multiple optimal paths, where the optimal paths correspond one-to-one to the task coordinate groups;

[0308] Complete multi-UAV navigation path planning based on multiple optimal paths.

[0309] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0310] Furthermore, if the modules / units integrated into the electronic device 1 are implemented as software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. The computer-readable storage medium may be volatile or non-volatile. For example, the computer-readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0311] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program. When the computer program is executed by a processor of an electronic device, the computer program can implement:

[0312] Acquire a three-dimensional mapping space, wherein the three-dimensional mapping space is in the shape of a cube and includes: a spatial rectangular coordinate system and a plurality of cells, wherein the cell is in the shape of a cube, a unit length of the spatial rectangular coordinate system is a side length of the cell, each cell in the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstruction cells;

[0313] Determine the current wind speed, current gust vector, current rainfall, and the UAV mission sequence, wherein the UAV mission sequence includes: K mission coordinate groups, and the mission coordinate group includes: starting coordinates and end coordinates;

[0314] Acquire multiple historical fault groups, where the historical fault groups include: historical fault coordinates and historical fault times;

[0315] Based on the preset initial pheromone concentration and multiple historical fault groups, the three-dimensional mapping space is negatively pheromone-marked to obtain a negatively marked space;

[0316] Extract the jth task coordinate group from the UAV task sequence, where the initial value of j is 1;

[0317] An ant colony simulation is performed on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative label space to obtain an updated coordinate group sequence set, wherein the updated coordinate group sequence set includes: multiple optimal path sequences;

[0318] Using the updated coordinate group sequence set as the path coordinate group sequence set, let J = j + 1, let J be j, return to the step of extracting the j-th task coordinate group from the UAV task sequence, until j = K, and use the updated coordinate group sequence set to generate multiple optimal paths, where the optimal paths correspond one-to-one to the task coordinate groups;

[0319] Complete multi-UAV navigation path planning based on multiple optimal paths.

[0320] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only exemplary, and actual implementations may have other division methods.

[0321] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0322] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0323] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0324] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A multi-UAV navigation path planning method based on multi-task parameters, characterized in that: The method comprises: Acquire a three-dimensional mapping space, wherein the three-dimensional mapping space is in the shape of a cube and includes: a spatial rectangular coordinate system and a plurality of cells, wherein the cell is in the shape of a cube, a unit length of the spatial rectangular coordinate system is a side length of the cell, each cell in the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstruction cells; Determine the current wind speed, current gust vector, current rainfall, and the UAV mission sequence, wherein the UAV mission sequence includes: K mission coordinate groups, and the mission coordinate group includes: starting coordinates and end coordinates; Acquire multiple historical fault groups, where the historical fault groups include historical fault coordinates and historical fault counts. The historical fault coordinates refer to the coordinates of cells corresponding to locations where UAV faults occurred in history in the three-dimensional mapping space, and the historical fault counts refer to the number of times the UAV faults occurred at the historical fault coordinates. Based on the preset initial pheromone concentration and multiple historical fault groups, the three-dimensional mapping space is negatively pheromone-marked to obtain a negatively marked space; Extract the jth task coordinate group from the UAV task sequence, where the initial value of j is 1; An ant colony simulation is performed on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set, and the negative label space to obtain an updated coordinate group sequence set, wherein the updated coordinate group sequence set includes: multiple optimal path sequences, wherein the pre-constructed path coordinate group sequence set is an empty set, and the optimization heuristic function is as follows: ； in, To optimize the heuristic function, For the distance, For close distance, is the environmental impact index, is the corner energy consumption index, is the mutual interference index, is the neighboring pheromone concentration, is the number of cycles, is the cycle threshold, where the corner energy consumption index reflects the energy consumed by the drone due to direction changes when it moves in reality according to the movement path of the virtual ants in the negatively marked space. The environmental impact index reflects the energy consumed by the drone due to the resistance of wind and rainfall in the environment when it moves in reality according to the movement path of the virtual ants in the negatively marked space. The mutual interference index reflects the interference risk caused by the overlap of the paths of other drones during the execution of the mission. The neighboring pheromone concentration is the pheromone concentration corresponding to the neighboring cells. The updated coordinate group sequence set is used as the path coordinate group sequence set, and Let J be j, return to the step of extracting the jth task coordinate group from the UAV task sequence until j=K, and use the updated coordinate group sequence set to generate multiple optimal paths, where the optimal paths correspond one-to-one to the task coordinate groups; Complete multi-UAV navigation path planning based on multiple optimal paths.

2. The multi-UAV navigation path planning method based on multi-task parameters according to claim 1, characterized in that: The negative pheromone marking of the three-dimensional mapping space based on the preset initial pheromone concentration and multiple historical fault groups to obtain the negative marking space includes: Perform the following operations on each of the multiple historical fault groups: Identify the fault cells in the three-dimensional mapping space based on the historical fault coordinates in the historical fault group; Identifying a plurality of dangerous cells in a three-dimensional mapping space based on the faulty cell, wherein the dangerous cells are adjacent to the faulty cell in the three-dimensional mapping space; Calculate the danger pheromone concentration based on the initial pheromone concentration and the number of historical failures in the historical failure group; Calculate the barrier pheromone concentration based on the danger pheromone concentration, where the barrier pheromone concentration is half of the danger pheromone concentration; Marking a fault cell based on the obstacle pheromone concentration to obtain an obstacle-marked cell, and marking each of a plurality of dangerous cells based on the danger pheromone concentration to obtain a plurality of danger-marked cells; Aggregating the obstacle marking cells and the plurality of danger marking cells to obtain a plurality of exclusion cells, wherein the exclusion cells are obstacle marking cells or danger marking cells; Eliminate multiple exclusion cells from multiple cells in the three-dimensional mapping space to obtain multiple basic cells; Each of the plurality of basic cells is marked based on the initial pheromone concentration to obtain a plurality of initially marked cells; Aggregating multiple exclusion cells and multiple initial mark cells to obtain multiple initial cells, wherein the initial cells are exclusion cells or initial mark cells; A negative label space is identified based on multiple initial cells and a three-dimensional mapping space.

3. The multi-UAV navigation path planning method based on multi-task parameters according to claim 2, characterized in that: The optimized ant colony simulation is performed on the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative label space to obtain an updated coordinate group sequence set, including: According to the starting coordinates in the j-th task coordinate group, the starting cell is identified in the negative marking space, and the ending coordinates in the j-th task coordinate group are marked as the ending coordinates; Generate multiple virtual ants, and perform the following operations on each of the multiple virtual ants: Place the virtual ant into the starting cell to obtain the starting cell; Identify multiple neighboring cells of the starting cell in the negatively marked space; For each of multiple neighboring cells, perform the following operations: Perform transfer strategy analysis on neighboring cells and starting cells based on current wind speed, current gust vector, current rainfall, optimization heuristic function, termination coordinates, preset number of loops, pre-built memory coordinate group sequence, preset start time and path coordinate group sequence set to obtain transfer comprehensive index and update time; Summarize the transfer comprehensive indexes and update times respectively to obtain multiple transfer comprehensive indexes and multiple update times, wherein the transfer comprehensive indexes and update times are in one-to-one correspondence with adjacent cells; Based on multiple transfer comprehensive indexes and multiple update times, multiple adjacent cells are probabilistically transferred to obtain the transferred cells and landing time; Move the virtual ant to the transfer cell, obtain the landing cell, and identify the landing coordinates of the landing cell in the negative mark space; Combine the landing point coordinates and the landing time to obtain a landing point coordinate group; Using the landing point coordinate group to update the memory coordinate group sequence, to obtain an updated memory group sequence; Determine whether the landing point coordinates are equal to the ending coordinates; If the landing point coordinates are not equal to the ending coordinates, the landing point cell is used as the starting cell, the updated memory group sequence is used as the memory coordinate group sequence, the landing time is used as the starting time, and the process returns to the step of identifying multiple neighboring cells of the starting cell in the negative mark space until the landing point coordinates are equal to the ending coordinates; If the landing point coordinates are equal to the ending coordinates, the updated memory group sequence is used as the target path sequence; Summarizing the target path sequences to obtain multiple target path sequences, wherein the target path sequences include: multiple target coordinate groups, and the target coordinate groups include: target coordinates and target time; Perform path analysis on multiple target path sequences to obtain the path overlap and the total time of multiple paths; Determine the path overlap with a preset path threshold and compare the number of cycles with a preset cycle threshold; If the path overlap is less than the path threshold and the number of cycles is less than the cycle threshold, pheromone optimization is performed on the negative marking space according to the total time of multiple paths and the multiple target path sequences to obtain a path marking space, the number of updates is calculated according to the number of cycles, the path marking space is used as the negative marking space, the number of updates is used as the number of cycles, and the step of generating multiple virtual ants is returned to until the path overlap is greater than or equal to the path threshold or the number of cycles is greater than or equal to the cycle threshold; If the path overlap is greater than or equal to the path threshold or the number of cycles is greater than or equal to the cycle threshold, the optimal path sequence is determined based on the total time of multiple paths and multiple target path sequences, and the path coordinate group sequence set is updated according to the optimal path sequence to obtain an updated coordinate group sequence set.

4. The multi-UAV navigation path planning method based on multi-task parameters according to claim 3, characterized in that: The transfer strategy analysis is performed on the neighboring cells and the starting cell according to the current wind speed, the current gust vector, the current rainfall, the optimization heuristic function, the end coordinates, the preset number of cycles, the pre-built memory coordinate group sequence, the preset start time and the path coordinate group sequence set to obtain the transfer comprehensive index and update time, including: Identify the neighboring coordinates and starting coordinates of the neighboring cells and starting cell in the negative label space respectively; Calculate the far distance based on the adjacent coordinates and the ending coordinates; Get the nearest distance based on the neighboring coordinates and the starting coordinates; Calculate the moving time based on the preset resolution conversion rate, the nearest distance and the preset average speed; Calculate the update time based on the moving time and the starting time, where the update time is the sum of the moving time and the starting time; Determine whether the adjacent cell is the obstacle cell, and if the adjacent cell is the obstacle cell, use the preset zero probability as the transfer comprehensive index; Otherwise, the movement vector is determined based on the neighboring coordinates and the starting coordinates; Identifying a last coordinate group in the memory coordinate group sequence, wherein the last coordinate group is the last memory coordinate group in the memory coordinate group sequence; Identifying an advance coordinate group in the memory coordinate group sequence based on the last coordinate group, wherein the last coordinate group is adjacent to the advance coordinate group in the memory coordinate group sequence and lags behind the advance coordinate group; Extract the advance coordinates from the advance coordinate group, extract the final coordinates from the final coordinate group, obtain the memory vector based on the advance coordinates and the final coordinates, and calculate the corner energy consumption index based on the memory vector and the movement vector. The calculation formula is as follows: ； in, is the corner energy consumption index, is the memory vector, is the movement vector, is the modulus of the motion vector, is the modulus of the memory vector, is the inverse cosine function, is the preset unit energy consumption value; The environmental impact index is calculated based on the movement vector, current wind speed, current gust vector, and current rainfall. The calculation formula is as follows: ； in, is the environmental impact index, and are the current rainfall and current wind speed, respectively. and are the current gust vector and the preset rise vector respectively, is the modulus of the motion vector, and are the preset lift resistance coefficient and the preset wind speed resistance coefficient, and are the modulus of the lift vector and the modulus of the current gust vector respectively; Identify neighboring pheromone concentrations of neighboring cells; Based on the update time and adjacent coordinates, the interference characteristics of the path coordinate group sequence set are analyzed to obtain the mutual interference index; The number of cycles, neighboring pheromone concentration, near distance, far distance, mutual interference index, corner energy consumption index and environmental impact index are substituted into the optimization heuristic function to obtain the transfer comprehensive index.

5. The multi-UAV navigation path planning method based on multi-task parameters according to claim 4, characterized in that: The interference feature analysis of the path coordinate group sequence set based on the update time and the adjacent coordinates to obtain the mutual interference index includes: Determine whether the path coordinate group sequence set is an empty set. If the path coordinate group sequence set is an empty set, use the preset initial interference index as the mutual interference index. Otherwise, perform the following operations on each path coordinate group sequence in the path coordinate group sequence set: identifying close coordinate groups in the path coordinate group sequence based on update time; Extracting close coordinates and close time from the close coordinate group; The mutual interference factor is calculated based on the update time, adjacent coordinates, close coordinates and close time. The calculation formula is as follows: ； in, is the mutual interference factor, and They are update time and close time respectively. and are the horizontal, vertical and vertical coordinates in the close coordinates respectively. and are the horizontal, vertical and vertical coordinates in the adjacent coordinates respectively, is the hyperbolic tangent function; The mutual interference factors are aggregated to obtain multiple mutual interference factors, and a mutual interference index is calculated according to the multiple mutual interference factors, wherein the mutual interference index is the sum of the multiple mutual interference factors.

6. The multi-UAV navigation path planning method based on multi-task parameters according to claim 5, characterized in that: The method of performing probabilistic transfer of multiple adjacent cells based on multiple transfer comprehensive indexes and multiple update times to obtain transfer cells and landing times includes: For each of multiple neighboring cells, perform the following operations: The transfer probability is calculated based on the transfer comprehensive index corresponding to the adjacent cells; Probabilistic selection is performed according to the transition probability corresponding to each adjacent cell among multiple adjacent cells to obtain a transition cell; The update time corresponding to the transferred cell in multiple update times is used as the landing time.

7. The multi-UAV navigation path planning method based on multi-task parameters according to claim 6, characterized in that: The path parsing of multiple target path sequences to obtain the path overlap and the total time of multiple paths includes: Confirm the number of paths for multiple target path sequences; For each of the multiple target path sequences, the following operations are performed: Extract the ath target coordinate group from the target path sequence, where the initial value of a is 1; Extract the ath coordinate from the ath target coordinate group, where the ath coordinate is the target coordinate in the ath target coordinate group; Summarizing the a-th coordinate to obtain multiple a-th coordinates, and extracting multiple reference horizontal coordinates, multiple reference vertical coordinates, and multiple reference vertical coordinates from the multiple a-th coordinates; Calculating abscissa variance according to a plurality of reference abscissas, wherein the abscissa variance is the variance of the plurality of reference abscissas; Obtaining a vertical coordinate variance and a vertical coordinate variance based on a plurality of reference vertical coordinates and a plurality of reference vertical coordinates respectively; Calculate the average coordinate variance according to the horizontal coordinate variance, the vertical coordinate variance, and the vertical coordinate variance, wherein the average coordinate variance is the average of the horizontal coordinate variance, the vertical coordinate variance, and the vertical coordinate variance; Let A = a + 1, with A as a, and return to the step of determining the number of paths in the plurality of target path sequences until A = M, where M is the number of paths; Summarize the average coordinate variances to obtain multiple average coordinate variances, and calculate the path coincidence degree based on the multiple average coordinate variances; The following operations are performed on each of the multiple target path sequences: extracting a plurality of target times from a plurality of target coordinate groups in a target path sequence; Calculate the total path time based on multiple target times, where the total path time is the sum of the multiple target times; Sum up the total path times to get the total times of multiple paths.

8. The multi-UAV navigation path planning method based on multi-task parameters according to claim 7, characterized in that: The pheromone optimization of the negative marking space according to the total time of multiple paths and the multiple target path sequences to obtain the path marking space includes: Target path sequences are sequentially extracted from multiple target path sequences, and the following operations are performed on the extracted target path sequences: Calculate the target pheromone increment based on the initial pheromone concentration and the total path time corresponding to the target path sequence in multiple path total times; extracting a plurality of target coordinates from a plurality of target coordinate groups in a target path sequence; For each of the multiple target coordinates, perform the following operations: Identify the target cell in the negative marker space based on the target coordinates; Identify the target pheromone concentration of the target cell, and calculate the replacement pheromone concentration based on the target pheromone concentration and the target pheromone increment; Based on the replacement pheromone concentration, the target cell is updated with pheromone to obtain a supplemented cell; Summarize the supplementary cells to obtain multiple supplementary cells; The negative mark space is updated based on the multiple supplementary cells to obtain a supplementary three-dimensional space; The supplemented three-dimensional space is used as a negative labeling space, and the step of sequentially extracting target path sequences from multiple target path sequences is returned to until all multiple target path sequences are extracted, and the supplemented three-dimensional space is used as a path labeling space.

9. A multi-UAV navigation path planning system based on multi-task parameters, characterized in that: The system comprises: a mapping space construction module, configured to obtain a three-dimensional mapping space, wherein the three-dimensional mapping space is in the shape of a cube and includes: a spatial rectangular coordinate system and a plurality of cells, wherein the cell is in the shape of a cube, a unit length of the spatial rectangular coordinate system is a side length of the cell, each of the plurality of cells has a unique corresponding three-dimensional coordinate in the spatial rectangular coordinate system, and the plurality of cells are composed of a plurality of blank cells and a plurality of obstruction cells; An environmental data labeling module is used to confirm the current wind speed, the current gust vector, the current rainfall, and the UAV mission sequence, wherein the UAV mission sequence includes: K mission coordinate groups, the mission coordinate group includes: starting coordinates and end coordinates, obtain multiple historical fault groups, wherein the historical fault group includes: historical fault coordinates and historical fault counts, wherein the historical fault coordinates refer to: the coordinates of the cell corresponding to the location where the UAV has historically failed in the three-dimensional mapping space, and the historical fault count refers to: the number of times the UAV has failed at the historical fault coordinates, and perform negative pheromone labeling on the three-dimensional mapping space based on a preset initial pheromone concentration and multiple historical fault groups to obtain a negative labeling space; The ant colony algorithm optimization module is used to extract the j-th task coordinate group from the UAV task sequence, where the initial value of j is 1, and perform an ant colony simulation to optimize the j-th task coordinate group based on the current wind speed, the current gust vector, the current rainfall, the pre-constructed optimization heuristic function, the pre-constructed path coordinate group sequence set and the negative label space to obtain an updated coordinate group sequence set, where the updated coordinate group sequence set includes: multiple optimal path sequences, where the pre-constructed path coordinate group sequence set is an empty set, and the optimization heuristic function is as follows: ； in, To optimize the heuristic function, For the distance, For close distance, is the environmental impact index, is the corner energy consumption index, is the mutual interference index, is the neighboring pheromone concentration, is the number of cycles, is the cycle threshold, where the corner energy consumption index reflects the energy consumed by the drone due to direction changes when it moves in reality according to the movement path of the virtual ants in the negatively marked space. The environmental impact index reflects the energy consumed by the drone due to the resistance of wind and rainfall in the environment when it moves in reality according to the movement path of the virtual ants in the negatively marked space. The mutual interference index reflects the interference risk caused by the overlap of the paths of other drones during the execution of the mission. The neighboring pheromone concentration is the pheromone concentration corresponding to the neighboring cells. The optimal path generation module is used to update the coordinate group sequence set as the path coordinate group sequence set, and , taking J as j, return to the step of extracting the jth task coordinate group from the UAV task sequence until j=K, and use the updated coordinate group sequence set to generate multiple optimal paths, where the optimal paths correspond one-to-one to the task coordinate groups, and complete the multi-UAV navigation path planning based on the multiple optimal paths.

Images