Transformation method and device for large-scale unmanned aerial vehicle configuration

Abstract

The invention discloses a transformation method and device for large-scale unmanned aerial vehicle configuration, and the method comprises the steps: dividing an unmanned aerial vehicle cluster of an original configuration into a plurality of first sub-regions based on the attribute of a first region corresponding to each unmanned aerial vehicle in the unmanned aerial vehicle cluster of the original configuration; dividing the unmanned aerial vehicle cluster of the target configuration into a plurality of second sub-regions based on a second region attribute corresponding to each unmanned aerial vehicle in the unmanned aerial vehicle cluster of the target configuration; performing region pairing on the plurality of first sub-regions and the plurality of second sub-regions to generate a region pairing result; based on a region pairing result, performing node pairing on the unmanned aerial vehicle of the original configuration and the unmanned aerial vehicle of the target configuration to obtain a plurality of paired nodes; and performing dynamic obstacle avoidance processing on each pairing node in the plurality of pairing nodes to generate a path plan of the unmanned aerial vehicle configuration. Thereby, it is possible to calculate the complexity of a path plan from O (N!) ) is reduced to O (MN2), the calculation time of path planning is effectively reduced, and the efficiency of path planning tasks is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and particularly relates to a transformation method and device for large-scale unmanned aircraft configurations. Background Art

[0002] With the rapid development of unmanned aircraft technology, unmanned aircraft have been widely used in military fields, film and television aerial photography, environmental monitoring, geographical mapping, power line inspection, environmental monitoring and other civilian fields. Facing increasingly complex task requirements, a single unmanned aircraft is difficult to complete tasks independently. Therefore, the form of multiple unmanned aircrafts collaborating to execute tasks has emerged. Unmanned aircraft path planning is the core of unmanned aircraft collaborative control. Therefore, studying the collaborative path planning of multiple unmanned aircrafts has very important significance.

[0003] The essence of multi-unmanned aircraft collaborative path planning is to coordinate the paths between multiple unmanned aircrafts to complete collaborative tasks. Existing path methods based on graph search and sampling have high efficiency and excellent global optimization performance in environmental search. However, in large-scale environments, such algorithms have problems of dynamic obstacle avoidance and communication. Path planning methods based on heuristic algorithms, such as ant colony optimization method, particle swarm optimization method, bee colony optimization method, genetic algorithm and simulated annealing algorithm, can effectively perform path planning and task allocation in complex, dynamic and multi-constrained environments, but have limitations such as large computational overhead and being easily trapped in local optima. Learning-based algorithms can adjust their own strategies by continuously learning new information in the environment, have strong adaptability, and are suitable for dynamic and uncertain environments. However, learning algorithms rely strongly on large-scale and high-quality training data. When the data is insufficient or inaccurate, the performance of the algorithm may drop significantly, and the algorithm interpretability is poor, which is not conducive to system debugging and optimization.

[0004] Existing multi-unmanned aircraft collaborative path planning algorithms mainly focus on dynamic obstacle avoidance and search coverage problems, and do not involve much in the field of multi-unmanned aircraft configuration transformation. When performing multi-unmanned aircraft configuration transformation, the method of manually assigning tasks by pre-specifying the positions of the target points of the unmanned aircraft does not consider constraint problems such as the shortest path, minimum energy or optimal time of task allocation. Traditional algorithms only consider path planning after task allocation. The method based on manual allocation is time-consuming and not suitable for the transformation of arbitrary configurations. Especially when the number of multi-unmanned aircraft formations increases, the difficulty of task allocation and path planning increases greatly. Summary of the Invention

[0005] The present invention provides a transformation method and device for large-scale unmanned aircraft configurations; the method can reduce the calculation time in the path planning process of converting an unmanned aircraft cluster from an original configuration to a target configuration and improve the path planning efficiency.

[0006] According to the first aspect of the embodiments of the present invention, a transformation method for a large-scale unmanned aircraft configuration is provided. The method includes: dividing the unmanned aircraft cluster of the original configuration into several first sub-regions based on the first regional attribute corresponding to each unmanned aircraft in the unmanned aircraft cluster of the original configuration; dividing the unmanned aircraft cluster of the target configuration into several second sub-regions based on the second regional attribute corresponding to each unmanned aircraft in the unmanned aircraft cluster of the target configuration; performing regional pairing on the several first sub-regions corresponding to the original configuration and the several second sub-regions corresponding to the target configuration to generate a regional pairing result; based on the regional pairing result, performing node pairing on the unmanned aircraft of the original configuration and the unmanned aircraft of the target configuration to obtain several paired nodes; performing dynamic obstacle avoidance processing on each of the several paired nodes to generate a path planning of the unmanned aircraft configuration.

[0007] Optionally, the method further includes: taking the centroid position of the unmanned aircraft cluster of the original configuration as the origin to construct a first coordinate system; taking the centroid position of the unmanned aircraft cluster of the target configuration as the origin to construct a second coordinate system; for any unmanned aircraft in the original configuration: determining the first local coordinate information of the unmanned aircraft in the first coordinate system based on the original coordinate of the unmanned aircraft in the world coordinate system and the origin coordinate corresponding to the first coordinate system; determining the first included angle formed between the projection of the unmanned aircraft on the oxy plane corresponding to the first coordinate system and the X axis based on the first local coordinate information; determining the first regional attribute of the unmanned aircraft based on the first included angle and the preset sub-region corresponding to the original configuration; for any unmanned aircraft in the target configuration: determining the second local coordinate information of the unmanned aircraft in the second coordinate system based on the original coordinate of the unmanned aircraft in the world coordinate system and the origin coordinate corresponding to the second coordinate system; determining the second included angle formed between the projection of the unmanned aircraft on the oxy plane corresponding to the second coordinate system and the X axis based on the second local coordinate information; determining the second regional attribute of the unmanned aircraft based on the second included angle and the preset sub-region corresponding to the target configuration.

[0008] Optionally, the method further includes: respectively optimizing the region attributes corresponding to each unmanned aerial vehicle in the first sub-region and the second sub-region; the optimizing the region attributes corresponding to each unmanned aerial vehicle in the first sub-region includes: for any target sub-region among the several first sub-regions: determining the centroid position corresponding to the target sub-region based on the centroid positions of each unmanned aerial vehicle in the world coordinate system in the target sub-region; for any unmanned aerial vehicle in the target sub-region: determining the first distance between the unmanned aerial vehicle and the target sub-region based on the centroid position of the unmanned aerial vehicle and the centroid position of the target sub-region; selecting at least two adjacent sub-regions corresponding to the target sub-region from the several first sub-regions; determining the first distance between the unmanned aerial vehicle and each of the adjacent sub-regions, obtaining at least two first distances; determining the sub-region corresponding to the minimum first distance among the at least three first distances as the region to which the unmanned aerial vehicle belongs, and updating the region attribute of the unmanned aerial vehicle based on the attribute of the selected region; re-calculating the centroid position corresponding to the target sub-region based on the updated region attribute of the unmanned aerial vehicle; repeating the above steps until the number of iterations meets a preset threshold or the centroid position corresponding to the target sub-region does not change, then ending the update step of the region attributes corresponding to each unmanned aerial vehicle in the target sub-region.

[0009] Optionally, the generating a region pairing result by performing region pairing on the several first sub-regions corresponding to the original configuration and the several second sub-regions corresponding to the target configuration includes: determining the distances between each first sub-region and each second sub-region based on the first centroid position corresponding to each first sub-region in the original configuration and the second centroid position corresponding to each second sub-region in the target configuration, obtaining a distance matrix; using the Hungarian pairing algorithm to perform region pairing on the distance matrix to generate a region pairing result.

[0010] Optionally, based on the area pairing result, pair the drones in the original configuration and the drones in the target configuration to obtain a number of paired nodes, including: for any paired area in the area pairing result: obtain the first sub-area and the second sub-area corresponding to the paired area; if the first sub-area and the second sub-area have the same number of drone nodes, then for any target node in the first sub-area: search for the paired node corresponding to the target node from the second sub-area based on the principle of the shortest path to obtain a number of paired nodes; if the number of drone nodes in the first sub-area is greater than the number of drone nodes in the second sub-area, then for any target node in the second sub-area: search for the paired node corresponding to the target node from the first sub-area based on the principle of the shortest path; for the unpaired nodes in the first sub-area: select the adjacent sub-areas adjacent to the second sub-area from several second sub-areas; search for the paired nodes corresponding to the unpaired nodes from the adjacent sub-areas based on the principle of the shortest path to obtain a number of paired nodes; if the number of drone nodes in the second sub-area is greater than the number of drone nodes in the first sub-area, then for any target node in the first sub-area: search for the paired drone corresponding to the target node from the second sub-area based on the principle of the shortest path; for the unpaired nodes in the second sub-area: select the adjacent sub-areas adjacent to the first sub-area from several first sub-areas; search for the paired nodes corresponding to the unpaired nodes from the adjacent sub-areas based on the principle of the shortest path to obtain a number of paired nodes.

[0011] Optionally, the method further includes: globally optimizing each paired node based on the principle of the shortest path of the paired node path, and outputting the optimal paired node; the globally optimizing each paired node based on the principle of the shortest path of the paired node path and outputting the optimal paired node includes: for any paired node among the several paired nodes: obtain the starting node corresponding to the original configuration in the paired node and the target node corresponding to the target configuration; based on the coordinate information corresponding to the starting node and the coordinate information corresponding to the target node, determine the ideal path of the drone corresponding to the paired node; sort the ideal paths of the drones corresponding to each of the several paired nodes in descending order of length to generate a sorting result; sequentially perform the drone node replacement operation on the paired nodes among the several paired nodes according to the sorting result; if the ideal path corresponding to the paired node after the replacement operation is less than the original paired node, then replace the original paired node with the paired node after the replacement; repeat the above steps until the total length of the ideal paths of the drones generated by the several paired nodes remains unchanged, then end the drone node replacement operation in the paired nodes and output the optimal paired node.

[0012] Optionally, performing dynamic obstacle avoidance processing on each of the several paired nodes to generate a path planning for the UAV configuration, including: obtaining several ideal paths based on the ideal paths of the UAVs corresponding to each of the several paired nodes; sorting the several ideal paths in ascending order of length, and sequentially numbering the starting nodes of the UAVs corresponding to each of the several paired nodes according to the sorting result; for any one of the several paired nodes: expanding search nodes for the starting node corresponding to the paired node; and searching for a node that satisfies the minimum path cost and conforms to the safety constraint from the search nodes as a path node; using the path node as the next starting node and continuing the search operation for path nodes until the UAV reaches the target node to end the search operation; generating the flight path of the UAV corresponding to the paired node; setting the path nodes in the search operation as temporary obstacles; and setting the target node as a fixed obstacle; where the safety constraint is used to indicate that the node is not a temporary roadblock or a fixed obstacle when performing the search operation at the current moment; sequentially performing the above path node search operation on each of the several paired nodes according to the numbering order until each UAV in the UAV cluster reaches the corresponding target node, generating a path planning for the UAV configuration.

[0013] According to the second aspect of the embodiments of the present invention, there is also provided a transformation device for a large-scale UAV configuration, the device including: a first sub-region module, configured to divide the UAV cluster of the original configuration into several first sub-regions based on the first region attribute corresponding to each UAV in the UAV cluster of the original configuration; a second sub-region module, configured to divide the UAV cluster of the target configuration into several second sub-regions based on the second region attribute corresponding to each UAV in the UAV cluster of the target configuration; a region pairing module, configured to perform region pairing on the several first sub-regions corresponding to the original configuration and the several second sub-regions corresponding to the target configuration to generate a region pairing result; a node pairing module, configured to pair the UAVs of the original configuration and the UAVs of the target configuration based on the region pairing result to obtain several paired nodes; a dynamic obstacle avoidance processing module, configured to perform dynamic obstacle avoidance processing on each of the several paired nodes to generate a path planning for the UAV configuration.

[0014] According to the third aspect of the embodiments of the present invention, there is also provided an electronic device, including: a processor; a memory for storing executable instructions executable by the processor; the processor is configured to read the executable instructions from the memory and execute the instructions to implement the method as described in the first aspect.

[0015] According to a fourth aspect of the embodiments of the present invention, there is also provided a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the method described in the first aspect is implemented.

[0016] The embodiments of the present invention provide a method and a device for transforming a large-scale unmanned aircraft configuration. The method includes: First, based on the first region attributes corresponding to each unmanned aircraft in the unmanned aircraft cluster of the original configuration, the unmanned aircraft cluster of the original configuration is divided into several first sub-regions; Second, based on the second region attributes corresponding to each unmanned aircraft in the unmanned aircraft cluster of the target configuration, the unmanned aircraft cluster of the target configuration is divided into several second sub-regions; Then, the several first sub-regions corresponding to the original configuration and the several second sub-regions corresponding to the target configuration are subjected to region pairing to generate a region pairing result; Finally, based on the region pairing result, the unmanned aircraft of the original configuration and the unmanned aircraft of the target configuration are subjected to node pairing to obtain several paired nodes; and dynamic obstacle avoidance processing is performed on each of the several paired nodes to generate a path planning of the unmanned aircraft configuration. The method for transforming a large-scale unmanned aircraft configuration in this embodiment adopts a sub-region strategy, and combines continuous steps of region attribute optimization, region pairing, node pairing, and paired node optimization, which can not only achieve the global optimal path planning result of the unmanned aircraft cluster, but also reduce the computational complexity from O(N!) to O(MN 2 ), M ≤ N; thereby effectively reducing the calculation time of path planning and improving the efficiency of the path planning task. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Some specific embodiments of the present invention will be described in detail hereinafter with reference to the drawings in an exemplary and non-limiting manner. The same reference numerals in the drawings denote the same or similar components or parts. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings:

[0018] Figure 1 is a flowchart of a method for transforming a large-scale unmanned aircraft configuration provided by an embodiment of the present invention;

[0019] Figure 2 is a flowchart of a method for transforming a large-scale unmanned aircraft configuration provided by another embodiment of the present invention;

[0020] Figure 3 is a structural diagram of a device for transforming a large-scale unmanned aircraft configuration provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] To make the objectives, features, and advantages of the present invention more obvious and understandable, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0022] The UAV swarm in the original configuration: Generally refers to the state presented by the UAV swarm when it is initially deployed or before executing a specific mission plan, which is naturally formed or presented according to the basic assembly method among UAVs. Generally, it is based on a simple networking and distribution method, without a special formation or structure design for a specific mission. The UAV swarm in the target configuration: It is a specific formation, structure, or layout state that the UAV swarm needs to achieve after careful planning and design according to specific mission requirements, environmental conditions, and tactical requirements, etc., in order to achieve the best mission execution effect.

[0023] This embodiment solves the path planning problem of the UAV swarm converting from the original configuration to the target configuration. As Figure 1 shown, it is a schematic flowchart of a method for transforming the configuration of a large-scale UAV provided by an embodiment of the present invention.

[0024] A method for transforming the configuration of a large-scale UAV includes at least the following steps:

[0025] S101, based on the first regional attribute corresponding to each UAV in the UAV swarm in the original configuration, divide the UAV swarm in the original configuration into several first sub-regions;

[0026] S102, based on the second regional attribute corresponding to each UAV in the UAV swarm in the target configuration, divide the UAV swarm in the target configuration into several second sub-regions;

[0027] S103, perform regional pairing on several first sub-regions corresponding to the original configuration and several second sub-regions corresponding to the target configuration to generate a regional pairing result;

[0028] S104, based on the regional pairing result, perform node pairing on the UAVs in the original configuration and the UAVs in the target configuration to obtain several paired nodes;

[0029] S105, perform dynamic obstacle avoidance processing on each paired node among several paired nodes to generate the path planning of the UAV configuration.

[0030] In S101 and S102, the first region attribute is used to indicate the region information of the drone located in the preset sub-region corresponding to the original configuration. The second region attribute is used to indicate the region information of the drone located in the preset sub-region corresponding to the target configuration. The region attribute corresponding to each drone in the drone cluster of the original configuration or the target configuration can be directly obtained, or can be obtained by the following method.

[0031] Exemplarily, taking the centroid position of the drone cluster of the original configuration as the origin, a first coordinate system is constructed; taking the centroid position of the drone cluster of the target configuration as the origin, a second coordinate system is constructed; for any drone in the original configuration: based on the original coordinates of the drone in the world coordinate system and the origin coordinates corresponding to the first coordinate system, determining the first local coordinate information of the drone corresponding to the first coordinate system; based on the first local coordinate information, determining a first included angle formed between the projection of the drone on the oxy plane corresponding to the first coordinate system and the X axis; based on the first included angle and the preset sub-region corresponding to the original configuration, determining the first region attribute of the drone. For any drone in the target configuration: based on the original coordinates of the drone in the world coordinate system and the origin coordinates corresponding to the second coordinate system, determining the second local coordinate information of the drone corresponding to the second coordinate system; based on the second local coordinate information, determining a second included angle formed between the projection of the drone on the oxy plane corresponding to the second coordinate system and the X axis; based on the second included angle and the preset sub-region corresponding to the target configuration, determining the second region attribute of the drone.

[0032] For example: based on the centroid positions of each drone in the original configuration s, calculate the centroid position c of the drone cluster of the original configuration s s ; based on the centroid positions of each drone in the target configuration g, calculate the centroid position c of the drone cluster of the target configuration g g . Taking the centroid position of the drone cluster of the original configuration as the origin, the x, y, and z axes are respectively parallel to the x, y, and z axes of the coordinate system where the drone position coordinates are located, and the positive directions remain unchanged, to construct a first coordinate system o s x s y s z s . Taking the centroid position of the drone cluster of the target configuration as the origin, the x, y, and z axes are respectively parallel to the x, y, and z axes of the coordinate system where the drone position coordinates are located, and the positive directions remain unchanged, to construct a second coordinate system o g x g y g z gDivide the UAV swarm in the original configuration and the target configuration into several small swarms with close numbers for subsequent path planning. The division rule for k small areas of the UAV swarm in the original configuration or the target configuration is to start from the positive x-axis direction, rotate counterclockwise, and divide into one area every (360 / k)°, with a total of k small areas divided; these k small areas are used as the preset sub-areas corresponding to the original configuration or the target configuration. Subtract the original coordinates of the UAVs in the original configuration from the origin coordinates (i.e., the centroid position c s coordinates) of the first coordinate system to obtain the first local coordinate information of the UAVs in the first coordinate system; then calculate the first angle α formed between the projection of the UAVs on the o s x s y s plane and the X-axis based on the arctangent function. Subtract the original coordinates of the UAVs in the target configuration from the origin coordinates (centroid position c g coordinates) of the second coordinate system to obtain the second local coordinate information of the UAVs in the second coordinate system; then calculate the second angle β formed between the projection of the UAVs on the o g x g y g plane and the X-axis based on the arctangent function. Then, convert both the first angle α and the second angle β to the range [0, 360) degrees; according to the magnitude of the converted first angle, map the UAVs to the preset sub-areas of the original configuration to obtain the first area attribute of the UAVs; according to the magnitude of the converted second angle, map the UAVs to the preset sub-areas of the target configuration to obtain the second area attribute of the UAVs. Thus, the UAV swarm can be effectively divided into multiple small areas, providing a basis for subsequent path planning and scheduling.

[0033] In S103, based on a preset rule or algorithm model, perform area pairing on the sub-areas of the original configuration and the target configuration to generate an area pairing result.

[0034] Exemplarily, based on the first centroid position corresponding to each first sub-area in the original configuration and the second centroid position corresponding to each second sub-area in the target configuration, determine the distances between each first sub-area and each second sub-area to obtain a distance matrix; use the Hungarian pairing algorithm to perform area pairing on the distance matrix to generate an area pairing result.

[0035] For example, for any first sub-region in the original configuration: Obtain the centroid positions of all the drones within this first sub-region; sum up and average the centroid positions of all the drones to obtain the first centroid position corresponding to the first sub-region. For any second sub-region in the target configuration: Obtain the centroid positions of all the drones within this second sub-region; sum up and average the centroid positions of all the drones to obtain the second centroid position corresponding to the second sub-region. Thus, the centroid positions of each sub-region in the original configuration and the target configuration in the global coordinate system are obtained. and

[0036] Calculate the distances between the centroids of each sub-region in the original configuration and each sub-region in the target configuration respectively to obtain the distance matrix D.

[0037]

[0038] wherein represents the distance between the centroid of the first sub-region s1 in the original configuration and the centroid of the second sub-region g1 in the target configuration; this distance can be calculated using the Euclidean distance formula.

[0039] Taking the distance matrix D as the input, use the Hungarian algorithm to complete the initial pairing of the sub-regions of the original configuration and the target configuration. The Hungarian algorithm can find the optimal region pairing within polynomial time, making the total distance after pairing the smallest. The output region pairing result includes the matching relationship of each pair of sub-regions.

[0040] Thus, it is possible to effectively provide a preliminary matching relationship for the sub-region pairing of the drones. This method utilizes the geometric characteristics of the centroid positions and combines the optimization ability of the Hungarian algorithm, thereby improving the efficiency of subsequent drone pairing and reducing the time required for optimization; furthermore, it lays a foundation for realizing efficient multi-drone collaborative operations.

[0041] In S104, based on the region pairing result, the UAVs in the original configuration and the UAVs in the target configuration are paired by nodes according to a preset rule or a model algorithm to obtain a number of paired nodes. Exemplarily, for any paired region in the region pairing result: obtain the first sub-region and the second sub-region corresponding to the paired region; if the first sub-region and the second sub-region have the same number of UAV nodes, then for any target node in the first sub-region: search for the paired node corresponding to the target node from the second sub-region based on the principle of the shortest path to obtain a number of paired nodes; if the number of UAV nodes in the first sub-region is greater than the number of UAV nodes in the second sub-region, then for any target node in the second sub-region: search for the paired node corresponding to the target node from the first sub-region based on the principle of the shortest path; for the unpaired nodes in the first sub-region: select the adjacent sub-region adjacent to the second sub-region from several second sub-regions; search for the paired node corresponding to the unpaired node from the adjacent sub-region based on the principle of the shortest path; obtain a number of paired nodes; if the number of UAV nodes in the second sub-region is greater than the number of UAV nodes in the first sub-region, then for any target node in the first sub-region: search for the paired UAV corresponding to the target node from the second sub-region based on the principle of the shortest path; for the unpaired nodes in the second sub-region: select the adjacent sub-region adjacent to the first sub-region from several first sub-regions; search for the paired node corresponding to the unpaired node from the adjacent sub-region based on the principle of the shortest path to obtain a number of paired nodes.

[0042] For example: for any paired region (s i , g j ): judge the number of UAVs in s i , g j in the paired region (s i , g j ). If the number of UAVs in (s i , g j ) is the same, then execute step S1; if the number of UAVs in s i is greater than the number of UAVs in g j , then execute steps S1 and S2; if the number of UAVs in s i is less than the number of UAVs in g j , then execute steps S1 and S3.

[0043] S1: Search and pair the UAVs in s i and gj in order according to the principle of the shortest path.

[0044] Use distance calculation algorithms (such as Euclidean distance or Manhattan distance) to evaluate the distance between the target nodes and the paired nodes within the paired area; and sort them in ascending order of distance and perform pairing in sequence. Alternatively, use the Hungarian algorithm to complete the initial pairing of drones between the target nodes and the paired nodes within the paired area. The Hungarian algorithm can find the optimal node pairing in polynomial time, minimizing the total distance after pairing. The output area pairing result includes the drone matching relationship of each pair of nodes.

[0045] S2:s i The unpaired drones within the set take turns to search for target nodes in other areas nearby for pairing in sequence. Each drone selects the unpaired target node closest to itself for pairing. Ensure that the paired drones do not participate in subsequent pairings.

[0046] S3:g j The unpaired drones within the set take turns to search for targets in other areas nearby for pairing in sequence. Adopt a similar strategy to select the closest unpaired drone for pairing; continue until all drones are paired or no suitable targets can be found anymore.

[0047] Thus, perform node pairing on the drones within the paired area to ensure that the drones in different areas can effectively find targets. This method can improve the pairing efficiency of drones within the paired area through the principle of the shortest path and takes into account the imbalance between the set and the number of internal drones in order to take corresponding remedial measures.

[0048] In S105, perform dynamic obstacle avoidance processing on each paired node among several paired nodes based on a preset rule or algorithm model to generate a path plan for the drone configuration.

[0049] Exemplarily, dynamic obstacle avoidance processing is performed on each of the several paired nodes to generate a path planning of the unmanned aircraft configuration; including: obtaining a plurality of ideal paths based on the ideal paths of the unmanned aircraft corresponding to each of the several paired nodes; sorting the plurality of ideal paths in ascending order of length, and sequentially numbering the starting nodes of the unmanned aircraft corresponding to each of the several paired nodes according to the sorting result; for any one of the several paired nodes: expanding search nodes for the starting node corresponding to the paired node; and searching for a node that satisfies the minimum path cost and conforms to the safety constraint from the search nodes as a path node; using the path node as the next starting node, and continuing the search operation for the path node until the unmanned aircraft reaches the target node to end the search operation; generating a flight path of the unmanned aircraft corresponding to the paired node; setting the path nodes in the search operation as temporary obstacles; and setting the target node as a fixed obstacle; wherein, the safety constraint is used to indicate that the node is not a temporary roadblock or a fixed obstacle when performing the search operation at the current moment; according to the numbering order, sequentially perform the above path node search operation on each of the several paired nodes until each unmanned aircraft in the unmanned aircraft cluster reaches the corresponding target node, generating a path planning of the unmanned aircraft configuration.

[0050] For example: S1: Calculate the initial estimated cost

[0051] For each unmanned aircraft, calculate the ideal path (i.e., the initial estimated cost) from its starting position to the target position using the Euclidean distance or other suitable distance calculation methods. Sort all the unmanned aircraft according to the ideal path length, and assign a unique number to each unmanned aircraft for subsequent processing.

[0052] S2: Expand search nodes

[0053] In the order of the unmanned aircraft numbers, expand the search nodes of each unmanned aircraft in turn: for each unmanned aircraft, use the A* algorithm to expand its adjacent nodes. The unmanned aircraft selects the path node with the minimum cost to move forward, and the path cost calculation includes the current path length and the estimated cost to reach the target.

[0054] S3: Traverse the next path point and handle collisions

[0055] For the next path node of each unmanned aircraft, perform the following operations:

[0056] Traverse the next path nodes of all the unmanned aircraft to check whether there is a collision risk (i.e., whether the path node is occupied by other unmanned aircraft or will cause a conflict). If a path node that does not meet the safety constraint is found, mark it as an obstacle to prevent the unmanned aircraft from selecting this path node. Re-plan the path based on the new state to ensure safety.

[0057] S4: Set up fixed obstacles

[0058] For the UAVs that have reached the target configuration, set the path nodes where they are located as fixed obstacles: This is done to prevent other UAVs from flying to this position subsequently, thus preventing collisions with the UAVs that have reached the target.

[0059] S5: Repeat path planning

[0060] Repeat steps S2 to S4 until all UAVs successfully reach their respective target positions: In each iteration, continuously check and update the status of path nodes.

[0061] Since the traditional A* algorithm can only avoid static obstacles and cannot handle the collision problem among multiple UAVs. The improved A* algorithm in this embodiment focuses on solving the dynamic obstacle avoidance problem during the flight of multiple UAVs. Its core logic is to sort and number the UAVs, expand the search nodes in sequence, continuously screen the path nodes, set the nodes that do not meet the safety constraints as obstacles, and at the same time set the nodes where the UAVs that have reached the target are located as fixed obstacles, and continuously iterate until all UAVs reach the target positions; thus, through the above steps, the A* algorithm is improved, making the algorithm not only able to avoid static obstacles, but also effectively handle the dynamic collision problem among multiple UAVs, greatly improving the safety and efficiency of the multi-UAV system flying in a complex environment.

[0062] The method for transforming the configuration of a large-scale UAV fleet in this embodiment adopts a sub-region strategy, and combines region pairing, node pairing, and dynamic obstacle avoidance processing steps, not only reducing the computational complexity during path planning, but also being able to effectively handle the collision problem during the flight of multiple UAVs, greatly improving the safety and efficiency of the multi-UAV system flying in a complex environment.

[0063] As Figure 2 shown, it is a schematic flowchart of a method for transforming the configuration of a large-scale UAV fleet provided by another embodiment of the present invention.

[0064] A method for transforming the configuration of a large-scale UAV fleet at least includes the following steps:

[0065] S201. Based on the first region attribute corresponding to each UAV in the UAV cluster of the original configuration, divide the UAV cluster of the original configuration into several first sub-regions;

[0066] S202. Based on the second region attribute corresponding to each UAV in the UAV cluster of the target configuration, divide the UAV cluster of the target configuration into several second sub-regions;

[0067] S203. Optimize the area attributes corresponding to each drone in the first sub-region and the second sub-region respectively, and obtain the optimized first sub-region and the optimized second sub-region.

[0068] S204. Pair the optimized first sub-regions and the optimized second sub-regions to generate a region pairing result.

[0069] S205. Based on the region pairing result, pair the drones in the original configuration and the drones in the target configuration to obtain a number of paired nodes.

[0070] S206. Based on the principle of the shortest path of the paired nodes, globally optimize each paired node among the number of paired nodes, and output the optimal paired nodes.

[0071] S207. Perform dynamic obstacle avoidance processing on each optimal paired node among the number of paired nodes to generate a path planning of the drone configuration.

[0072] It should be noted that the implementation processes of steps S201, S202, S204, S205, and S207 in this embodiment are similar to the implementation processes of S101, S102, S103, S104, and S105 in the Figure 1 embodiment respectively, and will not be repeated here.

[0073] In S203, optimize the area attributes of the drones in different configurations based on a preset rule or algorithm model.

[0074] Exemplarily, for any target sub-region among the several first sub-regions: determine the centroid position corresponding to the target sub-region based on the centroid positions of each drone in the target sub-region in the world coordinate system; for any drone in the target sub-region: determine the first distance between the drone and the target sub-region based on the centroid position of the drone and the centroid position of the target sub-region; select at least two adjacent sub-regions corresponding to the target sub-region from the several first sub-regions; determine the first distance between the drone and each of the adjacent sub-regions to obtain at least two first distances; determine the sub-region corresponding to the minimum first distance among the at least three first distances as the region to which the drone belongs, and update the area attribute of the drone based on the attributes of the selected region; based on the updated area attribute of the drone, recalculate the centroid position corresponding to the target sub-region; repeat the above steps until the number of iterations meets a preset threshold or the centroid position corresponding to the target sub-region does not change, and then end the update step of the area attributes corresponding to each drone in the target sub-region.

[0075] For example: S1. Calculate the centroid positions of each target sub-region.

[0076] S2, Calculate the distance: For any drone in the target sub-region, calculate the distance between the centroid of the drone and the centroid of the target sub-region where it is located, and the distance between the centroid of the drone and the centroid of the adjacent sub-region.

[0077] S3, Update the region attribute: According to the distance calculation result in S2, determine whether the drone should change the sub-region it belongs to. If the distance from the drone to the centroid of its current sub-region is greater than the distance to the centroid of an adjacent sub-region, update the sub-region attribute of the drone to the adjacent sub-region.

[0078] S4, Iterative update: Iterative process: Repeat steps S1 to S3 until either of the following conditions is met: 1. The set maximum number of iterations is reached. 2. The positions of the centroids of the target sub-regions no longer change during the iteration process (i.e., convergence).

[0079] S5, End condition: End: When the iteration is completed, the region attributes of all drones will be updated to the optimal state to ensure the successful execution of the path planning task.

[0080] In this embodiment, by continuously iterating and updating the region attributes of the drones, it is ensured that each drone is clearly assigned to the most suitable sub-region, thereby reducing mispairing in subsequent path planning and improving the success rate of the overall task.

[0081] In S206, exemplarily, for any one of the several paired nodes: Obtain the starting node corresponding to the original configuration and the target node corresponding to the target configuration in the paired node; Based on the coordinate information corresponding to the starting node and the coordinate information corresponding to the target node, determine the ideal path of the drone corresponding to the paired node; Sort the ideal paths of the drones corresponding to each of the several paired nodes in descending order of length to generate a sorting result; Perform a drone node replacement operation on the paired nodes in the several paired nodes in sequence according to the sorting result; If the ideal path corresponding to the paired node after the replacement operation is less than the original paired node, then replace the original paired node with the paired node after the replacement; Repeat the above steps until the total length of the ideal paths of the drones generated by the several paired nodes remains unchanged, then end the drone node replacement operation in the paired nodes and output the optimal paired nodes

[0082] For example: S1, Calculate the ideal path of the drone corresponding to the paired node;

[0083] S2, Traverse and sort the ideal path lengths; Traverse the ideal path lengths in all paired node sets and sort them in descending order of length, and record the index of each paired node for subsequent optimization. Generate an ordered list \(\{(L_{ij}, i, j)\}\), which contains the path length and the corresponding drone index.

[0084] S3. Optimize the paired node combinations; according to the optimization rules for the paired combinations of the starting node and the target node of the UAV, attempt to change the paired combination sequence of the UAV: select the paired combination (i, j) with the longest current path length for optimization; attempt to pair UAV i with other target points and calculate the new ideal path. Select the new paired combination. If the path length L′ of the new combination is less than the path length Lij of the original combination, then replace the original combination. Record the new paired combination and its path length.

[0085] S4: Repeat the optimization process. Repeat steps S1 to S3 until the total length of the ideal path of the UAV no longer changes.

[0086] In step S205 of this embodiment, the initial combination pairing of the UAVs in the original configuration and the target configuration is completed. However, the sub-region strategy is not an all - traversing strategy, and some of the UAV paired combinations are not the optimal combinations. Therefore, through the global optimization of the paired nodes in this embodiment, the influence of the maximum path length can be eliminated, thereby improving the overall efficiency and accuracy of node pairing, and further improving the transformation efficiency of the UAV cluster with different configurations.

[0087] The transformation method for large - scale UAV configurations in this embodiment adopts a sub - region strategy and combines consecutive steps of region attribute optimization, region pairing, node pairing, and paired node optimization. It can not only achieve the global optimal path planning result of the UAV cluster, but also reduce the computational complexity from O(N!) to O(MN 2 ), where M ≤ N; thus effectively reducing the computational time of path planning and improving the efficiency of the path planning task.

[0088] Next, a transformation method for large - scale UAV configurations provided in this embodiment will be described in detail in combination with a specific application scenario.

[0089] A transformation method for large - scale UAV configurations includes at least the following steps:

[0090] S1. Take the centroid position of the UAV cluster in the original configuration as the origin, and construct a first coordinate system; take the centroid position of the UAV cluster in the target configuration as the origin, and construct a second coordinate system. For any UAV in the original configuration: Based on the original coordinates of the UAV in the world coordinate system and the origin coordinates corresponding to the first coordinate system, determine the first local coordinate information of the UAV corresponding to the first coordinate system; based on the first local coordinate information, determine the first angle formed between the projection of the UAV on the oxy plane corresponding to the first coordinate system and the X-axis; based on the first angle and the preset sub-region corresponding to the original configuration, determine the first region attribute of the UAV. For any UAV in the target configuration: Based on the original coordinates of the UAV in the world coordinate system and the origin coordinates corresponding to the second coordinate system, determine the second local coordinate information of the UAV corresponding to the second coordinate system; based on the second local coordinate information, determine the second angle formed between the projection of the UAV on the oxy plane corresponding to the second coordinate system and the X-axis; based on the second angle and the preset sub-region corresponding to the target configuration, determine the second region attribute of the UAV.

[0091] S2. Based on the first region attribute corresponding to each UAV in the UAV cluster in the original configuration, divide the UAV cluster in the original configuration into several first sub-regions; based on the second region attribute corresponding to each UAV in the UAV cluster in the target configuration, divide the UAV cluster in the target configuration into several second sub-regions.

[0092] S3. For any target sub-region among several first sub-regions: Based on the centroid position of each UAV in the target sub-region in the world coordinate system, determine the centroid position corresponding to the target sub-region; for any UAV in the target sub-region: Based on the centroid position of the UAV and the centroid position of the target sub-region, determine the first distance from the UAV to the target sub-region; select at least two adjacent sub-regions corresponding to the target sub-region from several first sub-regions; determine the first distance from the UAV to each adjacent sub-region, and obtain at least two first distances; determine the sub-region corresponding to the minimum first distance among at least three first distances as the region to which the UAV belongs, and update the region attribute of the UAV based on the attribute of the selected region; based on the updated region attribute of the UAV, recalculate the centroid position corresponding to the target sub-region; repeat the above steps until the number of iterations meets the preset threshold or the centroid position corresponding to the target sub-region does not change, then end the update step of the region attribute corresponding to each UAV in the target sub-region.

[0093] S4. Based on the first centroid position corresponding to each first sub-region in the original configuration and the second centroid position corresponding to each second sub-region in the target configuration, determine the distances between each first sub-region and each second sub-region, and obtain a distance matrix; use the Hungarian matching algorithm to perform region matching on the distance matrix to generate a region matching result.

[0094] S5. For any paired region in the region pairing result: Obtain the first sub-region and the second sub-region corresponding to the paired region; if the first sub-region and the second sub-region have the same number of UAV nodes, then for any target node in the first sub-region: Search for the paired node corresponding to the target node from the second sub-region based on the principle of the shortest path, and obtain a number of paired nodes; if the number of UAV nodes in the first sub-region is greater than the number of UAV nodes in the second sub-region, then for any target node in the second sub-region: Search for the paired node corresponding to the target node from the first sub-region based on the principle of the shortest path; for the unpaired nodes in the first sub-region: Select the adjacent sub-region adjacent to the second sub-region from several second sub-regions; Search for the paired node corresponding to the unpaired node from the adjacent sub-region based on the principle of the shortest path; obtain a number of paired nodes; if the number of UAV nodes in the second sub-region is greater than the number of UAV nodes in the first sub-region, then for any target node in the first sub-region: Search for the paired UAV corresponding to the target node from the second sub-region based on the principle of the shortest path; for the unpaired nodes in the second sub-region: Select the adjacent sub-region adjacent to the first sub-region from several first sub-regions; Search for the paired node corresponding to the unpaired node from the adjacent sub-region based on the principle of the shortest path, and obtain a number of paired nodes.

[0095] S6. For any paired node among several paired nodes: Obtain the starting node corresponding to the original configuration and the target node corresponding to the target configuration in the paired node; Based on the coordinate information corresponding to the starting node and the coordinate information corresponding to the target node, determine the ideal path of the UAV corresponding to the paired node; Sort the ideal paths of the UAVs corresponding to each paired node among several paired nodes in descending order of length to generate a sorting result; Perform the UAV node replacement operation on the paired nodes among several paired nodes in sequence according to the sorting result; if the ideal path corresponding to the paired node after the replacement operation is less than the original paired node; then use the paired node after the replacement to replace the original paired node; Repeat the above steps until the total length of the ideal paths of the UAVs generated by several paired nodes remains unchanged, then end the UAV node replacement operation in the paired nodes and output the optimal paired nodes.

[0096] S7. Based on the ideal paths of the drones corresponding to each of the several optimal paired nodes, obtain several ideal paths; sort the several ideal paths in ascending order of length, and sequentially number the starting nodes of the drones corresponding to each paired node among the several paired nodes according to the sorting result; for any paired node among the several optimal paired nodes: expand search nodes for the starting node corresponding to the optimal paired node; and search for nodes that satisfy the minimum path cost and comply with the safety constraints from the search nodes as path nodes; use the path nodes as the next starting nodes and continue the search operation for path nodes until the drone reaches the target node to end the search operation; generate the flight path of the drone corresponding to the optimal paired node; set the path nodes in the search operation as temporary obstacles; and set the target node as a fixed obstacle; where the safety constraint is used to indicate that the node is not a temporary roadblock or a fixed obstacle when performing the search operation at the current moment; sequentially perform the above path node search operation for each of the several optimal paired nodes according to the numbered order until each drone in the drone cluster reaches the corresponding target node, and generate the path planning of the drone configuration.

[0097] As Figure 3 shown, it is a schematic structural diagram of a transformation device for a large-scale drone configuration provided by an embodiment of the present invention.

[0098] A transformation device for a large-scale drone configuration, the device 300 includes: a first sub-region module 301, configured to divide the drone cluster of the original configuration into several first sub-regions based on the first region attribute corresponding to each drone in the drone cluster of the original configuration; a second sub-region module 302, configured to divide the drone cluster of the target configuration into several second sub-regions based on the second region attribute corresponding to each drone in the drone cluster of the target configuration; a region pairing module 303, configured to perform region pairing on the several first sub-regions corresponding to the original configuration and the several second sub-regions corresponding to the target configuration to generate a region pairing result; a node pairing module 304, configured to pair the drones of the original configuration and the drones of the target configuration based on the region pairing result to obtain several paired nodes; a dynamic obstacle avoidance processing module 305, configured to perform dynamic obstacle avoidance processing on each of the several paired nodes to generate the path planning of the drone configuration.

[0099] In a preferred implementation manner of this embodiment, the device further includes: a first construction module, configured to use the centroid position of the drone cluster in the original configuration as the origin to construct a first coordinate system; a second construction module, configured to use the centroid position of the drone cluster in the target configuration as the origin to construct a second coordinate system; a first determination module, configured to, for any drone in the original configuration: based on the original coordinates of the drone in the world coordinate system and the origin coordinates corresponding to the first coordinate system, determine the first local coordinate information of the drone corresponding to the first coordinate system; based on the first local coordinate information, determine a first angle formed between the projection of the drone on the oxy plane corresponding to the first coordinate system and the X axis; based on the first angle and a preset sub-region corresponding to the original configuration, determine the first region attribute of the drone; a second determination module, configured to, for any drone in the target configuration: based on the original coordinates of the drone in the world coordinate system and the origin coordinates corresponding to the second coordinate system, determine the second local coordinate information of the drone corresponding to the second coordinate system; based on the second local coordinate information, determine a second angle formed between the projection of the drone on the oxy plane corresponding to the second coordinate system and the X axis; based on the second angle and a preset sub-region corresponding to the target configuration, determine the second region attribute of the drone.

[0100] In a preferred implementation manner of this embodiment, the device further includes: a region attribute optimization module, configured to perform optimization processing on the region attributes corresponding to each drone in the first sub-region and the second sub-region respectively. The region attribute optimization module includes: an optimization unit, configured to, for any target sub-region among the several first sub-regions: based on the centroid positions of each drone in the target sub-region in the world coordinate system, determine the centroid position corresponding to the target sub-region; for any drone in the target sub-region: based on the centroid position of the drone and the centroid position of the target sub-region, determine a first distance between the drone and the target sub-region; select at least two adjacent sub-regions corresponding to the target sub-region from the several first sub-regions; determine the first distances between the drone and each of the adjacent sub-regions, obtaining at least two first distances; determine the sub-region corresponding to the minimum first distance among the at least three first distances as the region to which the drone belongs, and update the region attribute of the drone based on the attribute of the selected region; based on the region attribute of the drone after update, recalculate the centroid position corresponding to the target sub-region; an iteration unit, configured to repeat the above steps until the number of iterations meets a preset threshold or the centroid position corresponding to the target sub-region does not change, and then end the update step of the region attributes corresponding to each drone in the target sub-region.

[0101] In a preferred embodiment of the present embodiment, the region pairing module includes: a determination unit, configured to determine the distances between each first sub-region and each second sub-region based on the first centroid positions corresponding to each first sub-region in the original configuration and the second centroid positions corresponding to each second sub-region in the target configuration, so as to obtain a distance matrix; a region pairing unit, configured to perform region pairing on the distance matrix by using the Hungarian pairing algorithm to generate a region pairing result.

[0102] In a preferred embodiment of the present embodiment, the node pairing module includes: an acquisition unit, configured to, for any paired region in the region pairing result: acquire the first sub-region and the second sub-region corresponding to the paired region; a first node pairing unit, configured to, if the first sub-region and the second sub-region have the same number of drone nodes, then for any target node in the first sub-region: search for the paired node corresponding to the target node from the second sub-region based on the principle of the shortest path to obtain a plurality of paired nodes; a first node pairing unit, configured to, if the number of drone nodes in the first sub-region is greater than the number of drone nodes in the second sub-region, then for any target node in the second sub-region: search for the paired node corresponding to the target node from the first sub-region based on the principle of the shortest path; for the unpaired nodes in the first sub-region: select the adjacent sub-region adjacent to the second sub-region from several second sub-regions; search for the paired node corresponding to the unpaired node from the adjacent sub-region based on the principle of the shortest path; to obtain a plurality of paired nodes; a third node pairing unit, configured to, if the number of drone nodes in the second sub-region is greater than the number of drone nodes in the first sub-region, then for any target node in the first sub-region: search for the paired drone corresponding to the target node from the second sub-region based on the principle of the shortest path; for the unpaired nodes in the second sub-region: select the adjacent sub-region adjacent to the first sub-region from several first sub-regions; search for the paired node corresponding to the unpaired node from the adjacent sub-region based on the principle of the shortest path to obtain a plurality of paired nodes.

[0103] In a preferred embodiment of the present embodiment, the device further includes: a global optimization module, configured to perform global optimization on each of the paired nodes based on the principle of the shortest path of the paired nodes, and output the optimal paired nodes. The global optimization module includes: a determination unit, configured to, for any one of the several paired nodes: obtain the starting node corresponding to the original configuration and the target node corresponding to the target configuration in the paired node; determine the ideal path of the unmanned aerial vehicle corresponding to the paired node based on the coordinate information corresponding to the starting node and the coordinate information corresponding to the target node; a replacement unit, configured to sort the ideal paths of the unmanned aerial vehicles corresponding to each of the several paired nodes in descending order of length to generate a sorting result; perform an unmanned aerial vehicle node replacement operation on the paired nodes in the several paired nodes in sequence according to the sorting result; if the ideal path corresponding to the paired node after the replacement operation is less than that of the original paired node; then replace the original paired node with the paired node after the replacement; a replacement operation unit, configured to repeat the above steps until the total length of the ideal paths of the unmanned aerial vehicles generated by the several paired nodes remains unchanged, then end the unmanned aerial vehicle node replacement operation in the paired nodes, and output the optimal paired nodes.

[0104] In a preferred embodiment of the present embodiment, the dynamic obstacle avoidance processing module includes: an acquisition unit, configured to obtain a plurality of ideal paths based on the ideal paths of the unmanned aerial vehicles corresponding to each of the several paired nodes; a numbering unit, configured to sort the plurality of ideal paths in ascending order of length, and number the starting nodes of the unmanned aerial vehicles corresponding to each of the several paired nodes in sequence according to the sorting result; a dynamic obstacle avoidance unit, configured to, for any one of the several paired nodes: expand search nodes for the starting node corresponding to the paired node; and search for a node that satisfies the minimum path cost and conforms to the safety constraint as a path node from the search nodes; use the path node as the next starting node, and continue the search operation for the path node until the unmanned aerial vehicle reaches the target node to end the search operation; generate the flight path of the unmanned aerial vehicle corresponding to the paired node; set the path node in the search operation as a temporary obstacle; and set the target node as a fixed obstacle; wherein, the safety constraint is used to indicate that the node is not a temporary roadblock or a fixed obstacle when performing the search operation at the current moment; a generation unit, configured to perform the above path node search operation on each of the several paired nodes in sequence according to the numbering order until each unmanned aerial vehicle in the unmanned aerial vehicle cluster reaches the corresponding target node, and generate the path planning of the unmanned aerial vehicle configuration.

[0105] The above device can execute a method for transforming a large-scale unmanned aircraft configuration provided by an embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing a method for transforming a large-scale unmanned aircraft configuration. For technical details not described in detail in this embodiment, reference may be made to a method for transforming a large-scale unmanned aircraft configuration provided by an embodiment of the present invention.

[0106] The present invention also provides an electronic device, including: a processor; a memory for storing executable instructions that can be executed by the processor; the processor is configured to read the executable instructions from the memory and execute the instructions to implement a method for transforming a large-scale unmanned aircraft configuration according to the present invention.

[0107] In addition to the above methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions that, when run by a processor, cause the processor to execute the steps in the methods according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0108] The computer program product can be written in any combination of one or more programming languages to write program code for performing the operations of the embodiments of the present application. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as an independent software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0109] In addition, an embodiment of the present application may also be a computer-readable storage medium, on which computer program instructions are stored, and the computer program instructions, when run by a processor, cause the processor to execute the steps in the methods according to the following embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0110] The computer-readable storage medium may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0111] The basic principles of the present application have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are only examples and not limitations. It cannot be considered that these advantages, benefits, effects, etc. are essential for each embodiment of the present application. In addition, the above-disclosed specific details are only for illustrative and understandable purposes and are not limitations. The above details do not limit the present application to necessarily adopt the above specific details for implementation.

[0112] It should also be noted that in the devices, equipment, and methods of the present application, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present application.

[0113] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0114] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0115] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the said claims.

Claims

1. A transformation method for large-scale unmanned machine configuration, characterized in that: include: Based on the first area attribute corresponding to each drone in the drone cluster of the original configuration, the drone cluster of the original configuration is divided into a plurality of first sub-areas; Based on the second area attribute corresponding to each drone in the drone cluster of the target configuration, the drone cluster of the target configuration is divided into a plurality of second sub-areas; Performing region pairing on a plurality of first sub-regions corresponding to the original configuration and a plurality of second sub-regions corresponding to the target configuration to generate a region pairing result; Based on the regional pairing results, pairing the drone of the original configuration with the drone of the target configuration to obtain a number of paired nodes; Dynamic obstacle avoidance processing is performed on each of the plurality of paired nodes to generate a path plan for the unmanned machine configuration.

2. The method according to claim 1, characterized in that: Also includes: The center of mass of the original configuration of the UAV cluster is taken as the origin to construct the first coordinate system; The center of mass of the target configuration UAV cluster is taken as the origin to construct the second coordinate system; For any drone in the original configuration: based on the original coordinates of the drone in the world coordinate system and the origin coordinates corresponding to the first coordinate system, determine the first local coordinate information of the drone corresponding to the first coordinate system; based on the first local coordinate information, determine the first angle formed between the oxy plane projection corresponding to the drone in the first coordinate system and the X-axis; based on the first angle and the preset sub-area corresponding to the original configuration, determine the first area attribute of the drone; For any UAV in the target configuration: based on the original coordinates of the UAV in the world coordinate system and the origin coordinates corresponding to the second coordinate system, determine the second local coordinate information of the UAV in the second coordinate system; based on the second local coordinate information, determine the second angle formed between the oxy plane projection corresponding to the UAV in the second coordinate system and the X-axis; based on the second angle and the preset sub-area corresponding to the target configuration, determine the second area attribute of the UAV.

3. The method according to claim 1, characterized in that: Before performing region pairing on the plurality of first sub-regions corresponding to the original configuration and the plurality of second sub-regions corresponding to the target configuration, the method further includes: optimizing the region attributes corresponding to each drone in the first sub-region and the second sub-region respectively; The optimizing process of the area attribute corresponding to each UAV in the first sub-area comprises: For any target sub-region among the several first sub-regions: based on the center of mass position of each drone in the target sub-region in the world coordinate system, determine the center of mass position corresponding to the target sub-region; for any drone in the target sub-region: based on the center of mass position of the drone and the center of mass position of the target sub-region, determine the first distance between the drone and the target sub-region; select at least two adjacent sub-regions corresponding to the target sub-region from the several first sub-regions; determine the first distance between the drone and each of the adjacent sub-regions to obtain at least two first distances; determine the sub-region corresponding to the smallest first distance among the at least three first distances as the region to which the drone belongs, and update the region attributes of the drone based on the attributes of the selected region; based on the updated region attributes of the drone, recalculate the center of mass position corresponding to the target sub-region; Repeat the above steps until the number of iterations meets the preset threshold or the centroid position corresponding to the target sub-region does not change, then end the step of updating the area attributes corresponding to each drone in the target sub-region.

4. The method according to claim 1, characterized in that: The performing region pairing on the plurality of first sub-regions corresponding to the original configuration and the plurality of second sub-regions corresponding to the target configuration to generate a region pairing result comprises: Based on the first centroid position corresponding to each first sub-region in the original configuration and the second centroid position corresponding to each second sub-region in the target configuration, determining the distance between each first sub-region and each second sub-region to obtain a distance matrix; The Hungarian pairing algorithm is used to perform region pairing on the distance matrix to generate a region pairing result.

5. The method according to claim 1, characterized in that: Based on the regional pairing result, the UAV of the original configuration and the UAV of the target configuration are paired to obtain a plurality of paired nodes; including: For any paired region in the region pairing results: obtaining a first sub-region and a second sub-region corresponding to the paired region; If the first sub-area and the second sub-area have the same number of drone nodes, then for any target node in the first sub-area: based on the shortest path principle, search for a pairing node corresponding to the target node in the second sub-area to obtain a number of pairing nodes; If the number of drone nodes in the first sub-area is greater than the number of drone nodes in the second sub-area, then for any target node in the second sub-area: search for a paired node corresponding to the target node from the first sub-area based on the shortest path principle; for an unpaired node in the first sub-area: select an adjacent sub-area adjacent to the second sub-area from a number of second sub-areas; search for a paired node corresponding to the unpaired node from the adjacent sub-areas based on the shortest path principle; and obtain a number of paired nodes; If the number of drone nodes in the second sub-area is greater than the number of drone nodes in the first sub-area, then for any target node in the first sub-area: search for a paired drone corresponding to the target node from the second sub-area based on the shortest path principle; for an unpaired node in the second sub-area: select an adjacent sub-area adjacent to the first sub-area from several first sub-areas; search for a paired node corresponding to the unpaired node from the adjacent sub-area based on the shortest path principle to obtain several paired nodes.

6. The method according to claim 5, characterized in that: Before performing dynamic obstacle avoidance processing on each of the plurality of paired nodes, the method further includes: performing global optimization on each paired node based on the principle of shortest paired node path, and outputting an optimal paired node; The method of globally optimizing each paired node based on the shortest paired node path principle and outputting the optimal paired node comprises: For any paired node among the plurality of paired nodes: obtaining a starting node corresponding to an original configuration in the paired node and a target node corresponding to the target configuration; determining an ideal path of the drone corresponding to the paired node based on coordinate information corresponding to the starting node and coordinate information corresponding to the target node; The ideal paths of the drones corresponding to each of the paired nodes in the plurality of paired nodes are sorted in descending order of length to generate a sorting result; according to the sorting result, the drone node replacement operation is performed on the paired nodes in the plurality of paired nodes in sequence; if the ideal path corresponding to the paired node after the replacement operation is smaller than the original paired node; the original paired node is replaced with the replaced paired node; Repeat the above steps until the total length of the ideal drone paths generated by the plurality of paired nodes remains unchanged, then end the drone node replacement operation in the paired nodes and output the optimal paired node.

7. The method according to claim 1, characterized in that: The method of performing dynamic obstacle avoidance processing on each of the plurality of paired nodes to generate a path plan for an unmanned machine configuration includes: Based on the ideal path of the drone corresponding to each of the paired nodes, obtaining a plurality of ideal paths; The plurality of ideal paths are sorted in ascending order of length, and the starting nodes of the drones corresponding to each of the plurality of paired nodes are numbered in sequence according to the sorting result; For any paired node among the plurality of paired nodes: expand the search node for the starting node corresponding to the paired node; search the node from the search node that satisfies the minimum path cost and meets the safety constraint as a path node; use the path node as the next starting node and continue the search operation for the path node until the drone reaches the target node; generate the drone flight path corresponding to the paired node; set the path node in the search operation as a temporary obstacle; and set the target node as a fixed obstacle; wherein the safety constraint is used to indicate that the node is not a temporary roadblock or a fixed obstacle when the search operation is performed at the current moment; According to the numbering sequence, the above-mentioned path node search operation is performed on each of the several paired nodes in turn until each drone in the drone cluster reaches the corresponding target node, thereby generating a path plan for the drone configuration.

8. A conversion device for large-scale unmanned machine configuration, characterized in that: include: A first sub-region module, configured to divide the drone cluster of the original configuration into a plurality of first sub-regions based on a first region attribute corresponding to each drone in the drone cluster of the original configuration; A second sub-region module, configured to divide the target configuration UAV cluster into a plurality of second sub-regions based on a second region attribute corresponding to each UAV in the target configuration UAV cluster; A region pairing module, used for performing region pairing on a plurality of first sub-regions corresponding to the original configuration and a plurality of second sub-regions corresponding to the target configuration, and generating a region pairing result; A node pairing module, used for pairing the UAV of the original configuration with the UAV of the target configuration based on the regional pairing result to obtain a plurality of paired nodes; The dynamic obstacle avoidance processing module is used to perform dynamic obstacle avoidance processing on each of the plurality of paired nodes to generate a path plan for the unmanned machine configuration.

9. An electronic device, comprising: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the instructions to implement the method according to any one of claims 1-7.

10. A computer readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images