Distributed multi-unmanned aerial vehicle motion planning method and system for electric power inspection

Abstract

The invention discloses a distributed multi-unmanned aerial vehicle motion planning method and system for electric power inspection. The method comprises the following steps: adding a collision avoidance strategy of flying to the right on the basis of a JPS algorithm; an improved JPS algorithm is combined with an SCA algorithm, and collision prevention of dynamic obstacles and static obstacles in the power inspection process of the unmanned aerial vehicle is achieved. According to the invention, on the global path planning level, on the basis of a traditional JPS method, a flight strategy which is close to the right relative to the current flight direction is selected, and an improved JPS algorithm and an SCA algorithm are combined and applied to a power inspection scene in which multiple unmanned aerial vehicles cooperate, so that collision prevention of dynamic obstacles and static obstacles in the power inspection process of the unmanned aerial vehicles can be realized; the problem of motion planning of the multiple unmanned aerial vehicles in an electric power inspection scene is successfully solved, the deadlock problem between the unmanned aerial vehicles and between the unmanned aerial vehicles and obstacles is reduced, and the challenge that the multiple unmanned aerial vehicles land in the aircraft nest at the same time is considered and overcome under the condition that the position of the aircraft nest is known.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicles, and in particular relates to a distributed multi-unmanned aerial vehicle motion planning method and system for power inspection. Background Art

[0002] Power inspection is the key to ensure the safety and stability of power equipment. The traditional manual inspection method has high risk factors, low inspection efficiency, and easy to miss inspections. Especially when inspecting overhead transmission lines facing mountains, power towers in mountain ravines, and related power equipment, the disadvantages of manual inspection are more prominent. Rotary-wing unmanned aerial vehicles (hereinafter referred to as drones) have the advantages of high flexibility and hovering operation. In recent years, with the development of drone technology, the use of drones instead of manual inspections has become a feasible technical solution. Motion planning is one of the key technologies for drones to perform power inspection tasks. It can ensure that drones avoid obstacles (such as mountains, trees, etc.) and no-fly zones (such as no-fly zones formed by electromagnetic field interference on drones) during flight, and complete power inspection tasks with the shortest flight path. However, the endurance of drones is usually poor, and the maximum flight time of general drones is about 45 minutes. Therefore, in a large-scale power inspection scenario, compared with a single drone, multiple drones working together to complete power inspection tasks have the advantages of more efficiency and accuracy, and can make up for the shortcomings of drone endurance.

[0003] However, although there are many motion planning methods for UAV collaboration, each method has corresponding shortcomings. Among them, although the centralized multi-UAV motion planning method (such as CBS algorithm, Conflict-Based Search) is simple in logic, the amount of calculation of the algorithm will increase exponentially with the increase of the scope of power inspection scenes and the number of UAVs, and it is not suitable for power inspection scenes with a large range or a large number of UAVs. Therefore, in actual power inspection scenes, distributed multi-UAV motion planning methods are more practical, such as the RVO (Reciprocal Velocity Obstacles) algorithm and its corresponding improved algorithm, which solves the deadlock problem between UAVs. However, due to the lack of global path guidance, in complex power inspection scenes, UAVs are still prone to deadlock when crossing non-convex obstacles. The introduction of a global path planning algorithm based on graph search (such as the A* algorithm or other corresponding improved algorithms) is an effective means to solve the deadlock problem when UAVs cross non-convex environments in power inspections.

[0004] However, in power inspection scenarios, the current distributed multi-UAV motion planning still affects the practical application of UAVs due to the following problems:

[0005] a. If the global path planning algorithm of graph search is introduced, in the power inspection scenario, drones are required to inspect overhead lines, transmission towers, and corresponding power equipment. Usually, the grid size of the power inspection space is smaller, which means that the search space of the A* algorithm and the corresponding improved algorithm in the power inspection scenario will become larger, resulting in an increase in the amount of calculation of the algorithm;

[0006] b. When inspecting overhead lines, drones may fly around large obstacles (such as mountains or dense forests). At this time, due to the lack of global path guidance, the drone may be deadlocked;

[0007] c. After completing the power inspection mission, all drones need to land in the nest to store or charge electricity. At this time, multiple drones may land in the nest at the same time. That is, after determining the precise location of the nest, the drones must ensure that they avoid collisions with each other while autonomously landing in the nest. Summary of the invention

[0008] In order to solve the technical problems existing in the prior art, the purpose of the present invention is to provide a distributed multi-UAV motion planning method and system for power inspection.

[0009] In order to achieve the above purpose and the above technical effect, the technical solution adopted by the present invention is:

[0010] A distributed multi-UAV motion planning method for power inspection includes the following steps:

[0011] Add a collision avoidance strategy of flying to the right based on the JPS algorithm;

[0012] The improved JPS algorithm is combined with the SCA algorithm to achieve collision avoidance of dynamic and static obstacles of UAVs during power inspection.

[0013] Further, the following steps are included:

[0014] 1) The three-dimensional space of the power inspection scene is rasterized, and the starting point of the drone is put into the openset set. Each time the path points added to the openset set are sorted from small to large according to the value of the cost f(x);

[0015] 2) Find all nodes x with minimum cost f(x) from the openset set min And put it into the set Ns min In the above example, we can judge the set Ns min Is the number of elements in 1? Give corresponding feedback and go to step 3) or step 4);

[0016] 3) In the rectangular coordinate system, the current node x of the drone will becurrent Points to the set Ns obtained in step 2) min The vector x formed by all the minimum cost nodes in min -x current Project to the XOY plane and update the current node x current , so as to get the latest current node;

[0017] 4) Update the current node x from the openset set current Delete and set the current node x current Add to the closedset collection;

[0018] 5) Determine the current node x current Whether it is the current inspection target position of the drone and give corresponding feedback, go to step 6) or end the global path planning;

[0019] 6) According to the current node x obtained current Determine the path-finding direction, determine the new node point that needs to be expanded according to the JPS algorithm, and then add it to the openset set in order of cost;

[0020] 7) Repeat step 2) to step 6) until the UAV obtains the collision avoidance path Path for the power inspection, wherein the collision avoidance path Path includes the collision avoidance paths of all inspection target locations;

[0021] 8) Each UAV executes steps 1) to 7) to obtain its own collision avoidance path Path, and inputs it and the map information of the power inspection scene into the SCA calculation module carried on the UAV;

[0022] 9) Each drone will set its current flight speed v i and the position p in the global coordinate system i Shared with all its neighboring drones, where i is the drone ID. Each drone calculates the flight speed at the next moment based on the perceived information and the received neighbor information, thereby achieving collision avoidance between drones and between drones and obstacles until the drone completes all inspection tasks.

[0023] Furthermore, the calculation formula of the cost f(x) is:

[0024] f(x)=g(x)+h(x)

[0025] Among them, g(x) is the path length of the UAV from the starting position to node x, and h(x) is the estimated path length of the UAV from node x to the inspection position.

[0026] Furthermore, in step 2), if the set Nsmin If the number of elements in is not 1, go to step 3); if the set Ns min If the number of elements in is 1, update the current node x current For the set Ns min The only minimum cost node x in min , then go to step 4).

[0027] Furthermore, in step 3), the vector x min -x current After projecting to the XOY plane, use Update the current node x current , thus getting the latest current node, where θ XOY (x min -x current ) is the angle between the vector projected onto the XOY plane and the positive direction of the X axis,

[0028] Furthermore, in step 5), if the current node x is determined current is the current inspection target position of the drone, then the global path planning ends; if the current node x current If it is not the current inspection target location of the drone, go to step 6).

[0029] Furthermore, if the drone has not completed all inspection tasks, it returns to step 9), and each drone sets its current flight speed v i and the position p in the global coordinate system i Shared with all its neighboring drones, the SCA calculation module deployed on each drone uses the SCA algorithm to calculate the flight speed at the next moment according to the location of obstacles within its own perception range, the no-fly zone formed by electromagnetic field interference, the speed and position of other neighboring drones, and the dynamic constraints of the drone until the drone completes all inspection tasks.

[0030] The present invention also discloses a distributed multi-UAV motion planning system for power inspection, comprising:

[0031] The spatial grid module is used to grid the three-dimensional space of the power inspection scene;

[0032] The data collection module is used to put the starting point of the drone into the openset set and sort the nodes added to the openset set from small to large according to the value of the cost f(x);

[0033] The first judgment module is used to find all nodes x with the minimum cost f(x) from the openset set. min And put it into the set Ns min In the above example, we can judge the set Nsmin Whether the number of elements in is 1;

[0034] The node update and processing module is used to obtain the updated current node, delete it from the openset set, and add it to the closedset set;

[0035] The second judgment module is used to judge whether the current node is the current inspection target location of the drone;

[0036] The node acquisition module is used to determine the path finding direction according to the current node, determine the new node point that needs to be expanded according to the JPS algorithm, and then add it to the openset set in the order of cost;

[0037] A collision avoidance path acquisition module is used to obtain collision avoidance paths for all inspection target locations;

[0038] The communication module is used to realize that each drone can transmit its current flight speed v i and the position p in the global coordinate system i Shared with all its neighbor drones;

[0039] The SCA calculation module is used to calculate the flight speed at the next moment using the SCA algorithm based on the obstacle positions within the drone's own perception range, the no-fly zone formed by electromagnetic field interference, the speed and position of other neighboring drones, and the drone's dynamic constraints.

[0040] The present invention also discloses an electronic device, comprising:

[0041] at least one processor;

[0042] a memory communicatively coupled to the at least one processor;

[0043] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the distributed multi-UAV motion planning method for power inspection as described above.

[0044] The present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned distributed multi-UAV motion planning method for power inspection.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1) The present invention discloses a distributed multi-UAV motion planning method and system for power inspection. At the global path planning level, considering that there is more than one path node with the minimum cost when a symmetric situation occurs, the present invention selects a flight strategy to the right relative to the current flight direction based on the traditional JPS method, which can effectively reduce the situation where UAVs flying in opposite directions meet and deadlock;

[0047] 2) The present invention discloses a distributed multi-UAV motion planning method and system for power inspection. Considering that after the UAVs complete the power inspection task, there may be multiple UAVs returning to the machine nest area at a similar time and the UAVs are prone to collision if the collision avoidance strategy is not considered during the landing process, the present invention can realize that multiple UAVs can safely land in the machine nest area at the same time without collision problems;

[0048] 3) The present invention discloses a distributed multi-UAV motion planning method and system for power inspection. At the global path planning level, the collision avoidance path for the UAVs to complete all inspection target locations has been obtained. Compared with the existing methods, the present invention can allow UAVs to fly safely through complex scenes such as woods and mountains, while avoiding the UAVs from being deadlocked in non-convex obstacle environments, with a higher safety factor. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is an example diagram of a symmetrical situation that a drone may encounter in a power inspection scenario of the present invention;

[0050] Figure 2 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0051] The present invention is described in detail below so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0052] A brief summary of one or more aspects is given below to provide a basic understanding of these aspects. This summary is not an exhaustive overview of all conceived aspects, and is neither intended to identify the key or critical elements of all aspects nor to define the scope of any or all aspects. Its only purpose is to give some concepts of one or more aspects in a simplified form as a prelude to a more detailed description that will be given later.

[0053] like Figure 1-2 As shown, a distributed multi-UAV motion planning method for power inspection can solve the deadlock problem of UAVs in most cases and improve the flight efficiency of UAVs when performing power inspection tasks. The method mainly includes the following steps:

[0054] First, we improve the JPS (Jump Points Search) algorithm. We add a collision avoidance strategy of flying to the right on the basis of the JPS algorithm, and avoid the deadlock problem of drones in symmetrical situations in global path planning. Symmetric situations such as Figure 1 As shown, and Respectively represent the path points To the inspection target point The symmetric situation we face and the path we get from the right-hand strategy, and Respectively represent the path points To the inspection target point The symmetrical situation faced and the path obtained by the right-hand driving strategy; secondly, the improved JPS algorithm is combined with the SCA (Shunted Collision Avoidance) algorithm to achieve the collision avoidance of dynamic obstacles and static obstacles of the UAV during the power inspection process. The SCA algorithm introduces the Dubins curve, so it can ensure the smoothness of the UAV flight trajectory.

[0055] The distributed multi-UAV motion planning method for power inspection of the present invention specifically includes the following steps:

[0056] 1) The three-dimensional space of the power inspection scene is rasterized, and the starting point of the drone is placed in the openset set, where the openset set is a set of key path points that the drone may fly through, and in subsequent steps, each path point added to the openset set (i.e., each grid position) is sorted from small to large according to the value of the cost f(x), f(x) = g(x) + h(x), g(x) is the path length of the drone from the starting position to node x, and h(x) is the estimated path length of the drone from node x to the inspection position (i.e., estimated as a straight path length).

[0057] 2) Find all nodes x with minimum cost f(x) from the openset set min And put it into the set Ns min In the above example, we can judge the set Ns min Whether the number of elements in is 1, such as can be determined by the len() function in Python, if the set Ns min If the number of elements in is not 1, go to step 3). If the set Ns min If the number of elements in is 1, update the current node x current For the set Ns min The only minimum cost node x in min , then go to step 4);

[0058] It should be noted that on a two-dimensional grid map, when a symmetrical situation occurs, there will be two nodes with the lowest cost, that is, when encountering an obstacle, the cost of the drone flying to the right or left of the obstacle to bypass the obstacle is the same. In three-dimensional space, there may be four nodes with the lowest cost, which is similar to the symmetrical situation in two-dimensional space, but expanded to three-dimensional space.

[0059] 3) In the rectangular coordinate system, the current node x of the drone will be current Points to Ns obtained in step 2) min The vector x formed by all the minimum cost nodes in min -x current (in, ) is projected onto the XOY plane, and the node with the smallest angle with the positive direction of the X axis is found in the two-dimensional vector after projection, and the node is updated to the latest current node x current , the calculation formula is shown in formula (1), where θ XOY (x min -x current ) is the angle between the vector projected onto the XOY plane and the positive direction of the X-axis;

[0060]

[0061] 4) Update the current node x from the openset set current Delete and set the current node x current Add to the closedset collection.

[0062] 5) By calculating the current node x current The distance from the target location of the current inspection of the drone is used to determine the current node x current Is it the current inspection target position of the drone? If it is the current inspection target position of the drone, the global path planning ends, otherwise continue with the following steps.

[0063] 6) According to the current node x obtained current Determine whether the pathfinding direction is a straight line or a diagonal direction, that is, if the expansion node and the current node x current If the formed vector is perpendicular or parallel to the XOY plane in the rectangular coordinate system, it is in the straight line direction, otherwise it is in the diagonal direction. Combined with the "two definitions and three rules" of the JPS algorithm, the new node point that needs to be expanded is determined, and then added to the openset set in order of cost.

[0064] 7) Repeat steps 2) to 6) until the UAV obtains the collision avoidance path Path for the power inspection. It should be noted that since the UAV often has more than one inspection target location when performing power inspection, at the global path planning level, the UAV needs to find the corresponding collision avoidance path according to the order of all inspections. Therefore, the final collision avoidance path Path of the UAV is a collision avoidance path that includes all inspection target locations.

[0065] 8) Each inspection UAV executes the above steps to obtain its own collision avoidance path, and inputs the obtained collision avoidance path and map information of the power inspection scene (mainly the position of obstacles in the global coordinate system) into the SCA calculation module deployed on each UAV.

[0066] 9) Each drone will set its current flight speed v i and the position p in the global coordinate system i Through the communication module, it is shared with all its neighboring drones, i is the drone ID number, so the SCA calculation module deployed on each drone uses the SCA algorithm to calculate the flight speed at the next moment according to the obstacle position within its own perception range, the no-fly zone formed by electromagnetic field interference, the speed and position of other neighboring drones, and the dynamic constraints of the drone, so as to achieve collision avoidance between drones and between drones and obstacles until the drone completes all inspection tasks. The steps of the SCA algorithm to calculate the speed of the drone at the next moment are as follows: the SCA algorithm first uses the current perception information to use the RVO algorithm to convert the speed atmosphere into a collision speed set and a feasible speed set, and selects the new speed closest to the ideal speed in the feasible speed set as the optimal feasible speed of the drone's flight at the next moment. If there is more than one optimal feasible speed, the optimal feasible speed on the right side of the drone's current flight direction is selected; and the ideal speed of the drone can be obtained by the Dubins curve between the current position of the drone and the inspection target position to obtain the ideal speed direction, and at the same time, the speed size is obtained according to the maximum flight speed of the drone, so as to calculate the ideal speed. If the drone has not completed all inspection tasks, return to step 9) and repeat the operation until the drone completes all inspection tasks.

[0067] The present invention also discloses a distributed multi-UAV motion planning system for power inspection, comprising:

[0068] The spatial grid module is used to grid the three-dimensional space of the power inspection scene;

[0069] The data collection module is used to put the starting point of the drone into the openset set and sort the nodes added to the openset set from small to large according to the value of the cost f(x);

[0070] The first judgment module is used to find all nodes x with the minimum cost f(x) from the openset set. min And put it into the set Ns min In the above example, we can judge the set Ns min Whether the number of elements in is 1;

[0071] The node update and processing module is used to obtain the updated current node, delete it from the openset set, and add it to the closedset set;

[0072] The second judgment module is used to judge whether the current node is the current inspection target location of the drone;

[0073] The node acquisition module is used to determine the path finding direction according to the current node, determine the new node point that needs to be expanded according to the JPS algorithm, and then add it to the openset set in the order of cost;

[0074] A collision avoidance path acquisition module is used to obtain collision avoidance paths for all inspection target locations;

[0075] The communication module is used to realize that each drone can transmit its current flight speed v i and the position p in the global coordinate system i Shared with all its neighbor drones;

[0076] The SCA calculation module is used to calculate the flight speed at the next moment using the SCA algorithm based on the obstacle positions within the drone's own perception range, the no-fly zone formed by electromagnetic field interference, the speed and position of other neighboring drones, and the drone's dynamic constraints.

[0077] The present invention also discloses an electronic device, comprising:

[0078] at least one processor;

[0079] a memory communicatively coupled to the at least one processor;

[0080] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the distributed multi-UAV motion planning method for power inspection as described above.

[0081] The present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned distributed multi-UAV motion planning method for power inspection.

[0082] Compared with the prior art, the present invention has at least the following advantages:

[0083] The present invention adds a collision avoidance strategy of flying to the right on the basis of the JPS algorithm, that is, when there are more than two nodes with the minimum cost in the extended nodes of the drone, the nodes on the right side of the current node direction or the nodes on the right side relative to the flight direction of the drone are preferentially searched for path finding. The key to its implementation is to find the minimum cost node on the right by projecting the vector from the current node of the drone to the newly extended minimum cost node in the three-dimensional space onto the XOY plane of the rectangular coordinate system. At the same time, considering the characteristics of drones in power inspection scenarios, the improved JPS algorithm is combined with the SCA algorithm for the first time and applied to the power inspection scenario of multi-drone collaboration, which can realize the collision avoidance of dynamic obstacles and static obstacles of drones in the power inspection process, successfully solve the motion planning problem of multiple drones in the power inspection scenario, reduce the deadlock problem between drones and drones, and between drones and obstacles, and consider and overcome the technical challenge of multiple drones landing on the machine nest at the same time while remembering the position of the machine nest.

[0084] Parts or structures not specifically described in the present invention may adopt existing technologies or existing products and will not be described in detail here.

[0085] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A distributed multi-UAV motion planning method for power inspection, characterized in that: The following steps are involved: Add a collision avoidance strategy of flying to the right based on the JPS algorithm; The improved JPS algorithm is combined with the SCA algorithm to achieve collision avoidance of dynamic and static obstacles of UAVs during power inspection.

2. A distributed multi-UAV motion planning method for power inspection according to claim 1, characterized in that: The following steps are involved: 1) The three-dimensional space of the power inspection scene is rasterized, and the starting point of the drone is put into the openset set. Each time the path points added to the openset set are sorted from small to large according to the value of the cost f(x); 2) Find all nodes x with minimum cost f(x) from the openset set min And put it into the set Ns min In the above example, we can judge the set Ns min Is the number of elements in 1? Give corresponding feedback and go to step 3) or step 4); 3) In the rectangular coordinate system, the current node x of the drone will be current Points to the set Ns obtained in step 2) min The vector x formed by all the minimum cost nodes in min -x current Project to the XOY plane and update the current node x current , so as to get the latest current node; 4) Update the current node x from the openset set current Delete and set the current node x current Add to the closedset collection; 5) Determine the current node x current Whether it is the current inspection target position of the drone and give corresponding feedback, go to step 6) or end the global path planning; 6) According to the current node x obtained current Determine the path-finding direction, determine the new node point that needs to be expanded according to the JPS algorithm, and then add it to the openset set in order of cost; 7) Repeat step 2) to step 6) until the UAV obtains the collision avoidance path Path for the power inspection, wherein the collision avoidance path Path includes the collision avoidance paths of all inspection target locations; 8) Each UAV executes steps 1) to 7) to obtain its own collision avoidance path Path, and inputs it and the map information of the power inspection scene into the SCA calculation module carried on the UAV; 9) Each drone will set its current flight speed v i and the position p in the global coordinate system i Shared with all its neighboring drones, where i is the drone ID. Each drone calculates the flight speed at the next moment based on the perceived information and the received neighbor information, thereby achieving collision avoidance between drones and between drones and obstacles until the drone completes all inspection tasks.

3. A distributed multi-UAV motion planning method for power inspection according to claim 2, characterized in that: The calculation formula of the cost f(x) is: f(x)=g(x)+h(x) Among them, g(x) is the path length of the UAV from the starting position to node x, and h(x) is the estimated path length of the UAV from node x to the inspection position.

4. A distributed multi-UAV motion planning method for power inspection according to claim 2, characterized in that: In step 2), if the set Ns min If the number of elements in is not 1, go to step 3); if the set Ns min If the number of elements in is 1, update the current node x current For the set Ns min The only minimum cost node x in min , then go to step 4).

5. A distributed multi-UAV motion planning method for power inspection according to claim 2, characterized in that: In step 3), vector x min -x current After projecting to the XOY plane, use Update the current node x current , thus getting the latest current node, where θ XOY (x min -x current ) is the angle between the vector projected onto the XOY plane and the positive direction of the X axis, N min .

6. A distributed multi-UAV motion planning method for power inspection according to claim 2, characterized in that: In step 5), if the current node x is determined current is the current inspection target position of the drone, then the global path planning ends; if the current node x current If it is not the current inspection target location of the drone, go to step 6).

7. A distributed multi-UAV motion planning method for power inspection according to claim 2, characterized in that: If the drone has not completed all inspection tasks, it returns to step 9), and each drone changes its current flight speed v i and the position p in the global coordinate system i Shared with all its neighboring drones, the SCA calculation module deployed on each drone uses the SCA algorithm to calculate the flight speed at the next moment according to the location of obstacles within its own perception range, the no-fly zone formed by electromagnetic field interference, the speed and position of other neighboring drones, and the dynamic constraints of the drone until the drone completes all inspection tasks.

8. A distributed multi-UAV motion planning system for power inspection, characterized in that: include: The spatial grid module is used to grid the three-dimensional space of the power inspection scene; The data collection module is used to put the starting point of the drone into the openset set and sort the nodes added to the openset set from small to large according to the value of the cost f(x); The first judgment module is used to find all nodes x with the minimum cost f(x) from the openset set. min And put it into the set Ns min In the above example, we can judge the set Ns min Whether the number of elements in is 1; The node update and processing module is used to obtain the updated current node, delete it from the openset set, and add it to the closedset set; The second judgment module is used to judge whether the current node is the current inspection target location of the drone; The node acquisition module is used to determine the path finding direction according to the current node, determine the new node point that needs to be expanded according to the JPS algorithm, and then add it to the openset set in the order of cost; A collision avoidance path acquisition module is used to obtain collision avoidance paths for all inspection target locations; The communication module is used to realize that each drone can transmit its current flight speed v i and the position p in the global coordinate system i Shared with all its neighbor drones; The SCA calculation module is used to calculate the flight speed at the next moment using the SCA algorithm based on the obstacle positions within the drone's own perception range, the no-fly zone formed by electromagnetic field interference, the speed and position of other neighboring drones, and the drone's dynamic constraints.

9. An electronic device, characterized in that: include: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute a distributed multi-UAV motion planning method for power inspection as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements a distributed multi-UAV motion planning method for power inspection as described in any one of claims 1 to 7.

Images