Inspection path planning method, equipment and product

Abstract

The embodiment of the invention provides a routing inspection track planning method, equipment and a product. The method comprises the following steps: determining a first initial track in an octree map, wherein the first initial track is a track extending from a routing inspection starting point to a routing inspection terminal point in the octree map; an extension point location of the first initial track is generated in the direction from the end point location on the first initial track to the inspection end point; according to the shortest path between the extension point location and other point locations on the first initial track, the path distance of the shortest path and the position relation between the shortest path and occupied nodes in the octree map, the first initial track is adjusted, and a first track is obtained; and according to the distance between a first track end point on the first track and a second track end point on a second track, generating an inspection track, the second track being a track extending from the inspection end point to the inspection starting point in the octree map. The method is used for achieving the effect of improving the path planning of the unmanned aerial vehicle.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular, to a patrol trajectory planning method, device, and product. Background Art

[0002] With the rapid development of artificial intelligence and unmanned aerial vehicle (UAV) technology, power patrol is gradually moving towards the unmanned era. UAVs play an increasingly important role in power facility patrols due to their flexibility and efficiency. However, in this field, there are still many challenges in key technical aspects such as patrol point setting and route planning, and innovative solutions are urgently needed to promote its large-scale application.

[0003] Currently, UAV power patrol mainly relies on pre-planned fixed routes for operation. Traditional patrol point setting is mostly completed through manual operation. In the setting method based on three-dimensional point cloud data, a large amount of point cloud data of power facilities is collected and used as the basis for subsequent analysis and processing. In terms of automatic route planning, the use of grid map forms is relatively common. It divides the environmental space into grids, compresses the environmental information by judging the state of each grid, and optimizes the route planning while meeting the safety flight constraints of the UAV. Especially when facing large-scale power equipment, this method can effectively improve the rate of planning optimization.

[0004] However, in the existing methods for UAV patrol trajectory planning, there are still problems with poor path planning effects. Summary of the Invention

[0005] Embodiments of this application provide a patrol trajectory planning method, device, and product to improve the effect of UAV path planning.

[0006] In a first aspect, embodiments of this application provide a patrol trajectory planning method, including:

[0007] Determine a first initial trajectory in the octree map, where the first initial trajectory is a trajectory extending from the patrol starting point to the patrol ending point in the octree map;

[0008] Generate an extension point of the first initial trajectory in the direction from the end point position on the first initial trajectory to the patrol ending point;

[0009] Adjust the first initial trajectory according to the shortest path between the extension point and other points on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map to obtain a first trajectory;

[0010] Generate a patrol trajectory according to the distance between the first trajectory end point on the first trajectory and the second trajectory end point on the second trajectory, where the second trajectory is a trajectory extending from the patrol ending point to the patrol starting point in the octree map.

[0011] In a possible implementation, the octree map is constructed based on the point cloud data of the space where the object to be inspected is located;

[0012] The types of occupied nodes in the octree map are different, and the resolutions of the occupied nodes are different.

[0013] In a possible implementation, the extension points of the first initial trajectory are generated in the direction from the end point position on the first initial trajectory to the inspection end point, including:

[0014] Generating random extension points in the direction from the end point position on the first initial trajectory to the inspection end point;

[0015] Determining the extension points in the random extension points according to the positional relationship between the position of the random path point and the occupied node, and the positional relationship between the path from the random path point to the end point position and the occupied node.

[0016] In a possible implementation, the first initial trajectory is adjusted according to the shortest path between the extension point and other points on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map to obtain the first trajectory, including:

[0017] Determining the shortest path from the extension point to other points on the first initial trajectory respectively, and the path distance of the shortest path;

[0018] According to the positional relationship between the shortest path and the occupied nodes in the octree map, determining the target other points among the other points, the shortest path distance between the target other points and the extension point, and the length from the inspection start point along the first initial trajectory to the target other points;

[0019] The first initial trajectory is adjusted according to the target other points among the other points, the shortest path distance between the target other points and the extension point, and the length from the inspection start point along the first initial trajectory to the target other points to obtain the first trajectory.

[0020] In a possible implementation, the angle between any two adjacent path segments on the first trajectory is greater than or equal to a preset angle threshold.

[0021] In a possible implementation, the inspection trajectory is generated according to the distance between the end point of the first trajectory and the end point of the second trajectory on the second trajectory, including:

[0022] If the distance between the end point of the first trajectory and the end point of the second trajectory on the second trajectory is less than a preset distance, an initial inspection trajectory is generated according to the first inspection trajectory and the second inspection trajectory;

[0023] Smooth the connections of path segments in the initial inspection trajectory to generate an inspection trajectory.

[0024] In a possible implementation, smoothing the connections of path segments in the initial inspection trajectory to generate an inspection trajectory includes:

[0025] Smooth the connections of path segments in the initial inspection trajectory to obtain a smooth inspection trajectory;

[0026] Optimize the smooth inspection trajectory according to the preset inspection speed condition and inspection acceleration condition to determine the inspection trajectory.

[0027] In a possible implementation, the method further includes:

[0028] If the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory is greater than or equal to the preset distance, then use the first trajectory as the first initial trajectory and re - execute the step of determining the first initial trajectory in the octree map until the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory is less than the preset distance to generate an inspection trajectory.

[0029] In a second aspect, an inspection trajectory planning device provided by an embodiment of the present application includes:

[0030] A determination module, configured to determine a first initial trajectory in the octree map, where the first initial trajectory is a trajectory extending from the inspection start point to the inspection end point direction in the octree map;

[0031] A point - position generation module, configured to generate extended point - positions of the first initial trajectory in the direction from the end - point position on the first initial trajectory to the inspection end point;

[0032] An adjustment module, configured to adjust the first initial trajectory according to the shortest path between the extended point - positions and other point - positions on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map to obtain a first trajectory;

[0033] A trajectory generation module, configured to generate an inspection trajectory according to the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory, where the second trajectory is a trajectory extending from the inspection end point to the inspection start point direction in the octree map.

[0034] In a third aspect, an inspection trajectory planning device provided by an embodiment of the present application includes: a memory, a processor;

[0035] The memory stores computer - executable instructions;

[0036] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementation manners of the first aspect.

[0037] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the above first aspect and / or various possible implementation manners of the first aspect.

[0038] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the above first aspect and / or various possible implementation manners of the first aspect.

[0039] An inspection trajectory planning method, device, and product provided by an embodiment of the present application can initially construct a basic path by determining a first initial trajectory extending from the inspection starting point to the inspection ending point, providing a direction framework for subsequent planning. Generating extension points and adjusting the first initial trajectory based on the shortest path, path distance between the extension points and other points on the first initial trajectory, and the position relationship with occupied nodes can effectively avoid obstacles and optimize the path length, improving the path safety and efficiency. At the same time, by combining a second trajectory extending from the inspection ending point to the inspection starting point and generating the final inspection trajectory quickly according to the distance between the two trajectory end points, the UAV path planning is made more comprehensive and reasonable, thus achieving the effect of improving the UAV path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings here are incorporated into the description and form a part of this description, showing embodiments consistent with the present application, and are used together with the description to explain the principles of the present application.

[0041] Figure 1 It is a schematic diagram of the scenario of the inspection trajectory planning method provided by the present application;

[0042] Figure 2 It is a flowchart of the inspection trajectory planning method provided by the present application Figure 1 ;

[0043] Figure 2a It is a schematic diagram of generating an inspection trajectory provided by an embodiment of the present application;

[0044] Figure 3 It is a flowchart of the inspection trajectory planning method provided by the present application Figure 2 ;

[0045] Figure 3a It is a flowchart of the inspection trajectory planning method provided by the present application Figure 2 ;

[0046] Figure 4Structural schematic diagram of the inspection trajectory planning device provided by the present application;

[0047] Figure 5 Structural schematic diagram of the inspection trajectory planning equipment provided by the present application.

[0048] Through the above-mentioned drawings, specific embodiments of the present application have been shown, and there will be more detailed descriptions hereinafter. These drawings and the written description are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed implementation manners

[0049] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0050] First, the terms involved in the present application are explained:

[0051] An octree is a hierarchical data structure for three-dimensional space subdivision. Each node in the octree represents the space contained in a cubic volume (usually called a voxel). The voxel is recursively subdivided into 8 sub-voxels until a given minimum voxel size is reached. The minimum voxel size determines the resolution of the octree.

[0052] In the field of UAV power inspection, the prior art covers multiple aspects. In terms of flight path planning, there are fixed flight path methods that rely on manually setting inspection points, attempts to set inspection points based on three-dimensional point cloud data, and occupancy grid maps used in automatic flight path planning. Path planning algorithms mostly stay at the theoretical level, mainly focusing on obstacle avoidance algorithms and mostly in the function simulation stage. In terms of environmental perception and map construction, methods such as three-dimensional point clouds, elevation maps, and multi-level surface maps are used to simulate the 3D environment. Voxel sampling is commonly used for point cloud processing, and there are also attempts to use octrees to convert the occupancy grid mapping from 2D to 3D for spatial modeling.

[0053] However, there are still many deficiencies in the current technology. When setting inspection points, there are many problems with the manual method, and three-dimensional point cloud data has low efficiency and poor reliability due to high memory occupancy. If the point cloud data is directly gridified in automatic flight path planning, calculation redundancy and detail loss are likely to occur. Path planning algorithms have strict requirements for the height information of the flight environment and poor engineering applications, and the obstacle avoidance strategy also affects the integrity of the inspection. In environmental perception and map construction, methods such as three-dimensional point clouds and elevation maps have their respective defects, such as low memory efficiency and inability to represent unmapped areas, and it is also difficult to balance the resolution and memory occupancy in voxel sampling.

[0054] An embodiment of the present application provides an inspection trajectory planning method. In an octree map, an extended point is generated in the direction from the end point position on the first initial trajectory to the inspection end point. The first initial trajectory is adjusted to obtain a first trajectory based on the shortest path, path distance between the extended point and other points on the first initial trajectory, and the positional relationship with occupied nodes. At the same time, a second trajectory is generated in the direction from the inspection end point to the inspection start point. Then, by introducing an adaptive multi-resolution 3D grid, using its characteristics of reducing computational redundancy and retaining device details, and using the octree structure to delay the initialization of the map volume and perform different-level cutting as a multi-resolution representation, technical problems such as many problems in the traditional inspection point setting method, high memory occupancy and easy loss of details when setting inspection points based on three-dimensional point cloud data, insufficient representation ability and difficulty in balancing memory occupancy and resolution in the environmental perception and map construction method, and reasonable planning and connection of the first trajectory and the second trajectory are solved.

[0055] Figure 1 is a schematic diagram of the scenario of the inspection trajectory planning method provided by the present application. As Figure 1 shown, the specific execution entity of the present application can be an inspection trajectory planning system, and the inspection trajectory planning system can be a server. Among them, the server can be a device such as a mobile phone, a tablet computer, or a computer. In this embodiment, there is no special limitation on the implementation manner of the execution entity, as long as the execution entity can determine the first initial trajectory in the octree map, the first initial trajectory is a trajectory extending from the inspection start point to the inspection end point direction in the octree map; generate an extended point of the first initial trajectory in the direction from the end point position on the first initial trajectory to the inspection end point; adjust the first initial trajectory according to the shortest path, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map to obtain a first trajectory; generate an inspection trajectory according to the distance between the first trajectory end point on the first trajectory and the second trajectory end point on the second trajectory, and the second trajectory is a trajectory extending from the inspection end point to the inspection start point direction in the octree map. That's it.

[0056] The following uses specific embodiments to describe in detail the technical solutions of the present application and how the technical solutions of the present application solve the above technical problems. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The following will describe the embodiments of the present application in conjunction with the drawings.

[0057] Figure 2 is a flow chart of the inspection trajectory planning method provided by the present application Figure 1 As Figure 2 shown, the method includes:

[0058] S201. Determine the first initial trajectory in the octree map. The first initial trajectory is the trajectory extending from the inspection starting point to the inspection ending point in the octree map.

[0059] Among them, the octree map can be based on the octree data structure. Its core principle is to recursively divide the three-dimensional space. Starting from a root node representing the entire space, each node corresponds to a cubic space area, which is evenly divided into eight sub-cubic spaces. These eight sub-spaces respectively correspond to the eight child nodes of the octree. Through continuous subdivision, an accurate description of the three-dimensional space is achieved.

[0060] The octree map includes nodes and node attributes. Among them, the node is the basic component unit of the octree map, including the root node, internal nodes, and leaf nodes. The root node represents the entire three-dimensional space. The internal nodes are the nodes in the middle division process, and the leaf nodes are the nodes that are divided to the bottom layer and no longer subdivided. Each node has a specific spatial range and attribute information. The node attribute can refer to the attribute of the node, which can include the spatial position attribute for determining the specific coordinates of the node in the three-dimensional space, the state attribute that can be divided into occupied, free, and undetermined states, and can also include other information related to the represented space, such as environmental data such as temperature and humidity.

[0061] When constructing the octree map, the spatial range to be modeled can be determined and used as the root node of the octree. Then, according to the distribution of objects in the space, the root node is divided according to certain rules. If there are objects in the spatial area represented by a certain node and the size of this area exceeds the preset accuracy requirement, it is continued to be divided into eight child nodes. Repeat this process until the spatial area represented by each leaf node meets the accuracy requirement or only contains a single state (such as completely occupied or completely free).

[0062] In the embodiment of the present application, the octree can be used to model Boolean attributes, which is reflected as the modeling of volume occupancy in the occupancy map planning. If it is detected that a certain voxel is occupied, the corresponding node in the octree is initialized, and the uninitialized nodes are regarded as free, so as to represent the occupancy state of the space with Boolean values.

[0063] Using the Boolean occupancy state or discrete labels, a compact representation of the octree can be achieved. When the states of all child nodes of a node are the same, whether occupied or free, these child nodes can be pruned, thus greatly reducing the number of nodes to be maintained in the tree and improving the storage and processing efficiency.

[0064] In the embodiment of the present application, the octree map can be the octree map of the space where the object to be inspected is located. Among them:

[0065] The inspection object can refer to power equipment that needs to be inspected, such as wires, cables, and other equipment.

[0066] The space where the object to be inspected is located can refer to a three-dimensional space area centered on the power equipment, including a certain range around it. This space not only covers the spatial positions occupied by power equipment such as wires and cables themselves, but also includes the connection parts between the equipment, the space between the equipment and the support structures (such as utility poles, towers, etc.), and the space within a certain safety distance range around the equipment that needs to be considered to ensure safe inspection and normal operation of the equipment.

[0067] For example, for overhead wires, the space where the object to be inspected is located not only includes the space occupied by the thickness of the wire itself, but also includes the spatial range involved in the sag of the wire between the poles, and a cylindrical space area centered on the wire and extending a certain distance around (such as the safety distance required considering possible wind deflection, discharge, etc.). For cables, it may refer to the internal space of the pipeline where the cables are laid, and a certain range of space around the pipeline demarcated for the convenience of maintenance, prevention of external interference, and other factors.

[0068] In addition, this space can also have an interaction relationship with other relevant facilities or environmental factors, such as nearby buildings, trees, roads, etc. When constructing the octree map, these factors need to be taken into consideration to more comprehensively and accurately reflect the actual spatial environment where the object to be inspected is located, provide accurate spatial information for inspection equipment such as drones, and thus achieve efficient and safe inspection tasks.

[0069] The first initial trajectory in the octree map can refer to a continuous path initially planned based on the three-dimensional spatial information constructed by the octree, starting from the inspection starting point and following the general direction pointing to the inspection end point. This initial trajectory can be the path during the planning process.

[0070] The inspection starting point can refer to the spatial position where the inspection task starts, which can be the entrance of the power facility concentration area or a specific exit of the substation, etc.; the inspection end point can refer to the spatial position where the task ends, such as the regional boundary point or near the monitoring site.

[0071] Among them, in the embodiments of the present application, the octree map is constructed according to the point cloud data of the space where the object to be inspected is located;

[0072] The types of occupied nodes in the octree map are different, and the resolutions of the occupied nodes are different.

[0073] Among them, when constructing an octree map from the point cloud data of the space where the object to be inspected is located, for the large-scale power transmission line point cloud, methods such as SOR (Statistical Outlier Removal) filtering are first used to filter out noise points and outliers, eliminate environmental interference, and improve the accuracy and reliability of the point cloud. Then, power facilities and their characteristics are identified, and a classification algorithm is used to classify the point cloud into categories such as plants, ground, buildings, and equipment.

[0074] Based on the overall range of the point cloud data, determine the three-dimensional space corresponding to the root node of the octree to ensure that it can cover all the point cloud data. Subsequently, create the root node in memory and associate it with the spatial range.

[0075] Enter the node division and construction link, traverse each point in the point cloud data one by one to determine whether it is within the current node space. For nodes containing multiple points, divide them into eight sub-nodes according to the octree rule, calculate the spatial range of each sub-node and allocate the corresponding point cloud data, and repeat this operation for the sub-nodes until the preset stop condition is met, such as the number of point clouds in the node is lower than the threshold or the spatial size is smaller than the preset value.

[0076] After the division is completed, assign attributes to the nodes. Determine the occupancy status based on the point cloud data in the node: "occupied" if there is data, "idle" if there is no data and both the parent node and ancestor nodes are "idle", and "undetermined" if the status is unclear. Attributes such as the average height and temperature of the node can also be assigned as needed, and relevant information is obtained from the point cloud or other sensors.

[0077] Finally, carry out optimization and post-processing. Check adjacent leaf nodes. If the occupancy status is the same and meets the merging conditions, merge them to reduce the number of nodes and improve the storage and processing efficiency; for small holes formed by missing point clouds, fill them according to the status of surrounding nodes to make the octree map more complete and accurate.

[0078] S202: Generate the extended points of the first initial trajectory in the direction from the end point position on the first initial trajectory to the inspection end point.

[0079] Among them, the end point position on the first initial trajectory can refer to the end position point of the first initial trajectory preliminarily planned.

[0080] The direction from the end point position to the inspection end point can refer to the vector direction formed by starting from the end point position of the first initial trajectory and pointing to the final target position (i.e., the inspection end point) of the inspection task in the three-dimensional space defined by the octree map.

[0081] In some embodiments, when generating the extended points, in the three-dimensional space coordinate system of the octree map, the coordinates of the first initial trajectory end point and the inspection end point can be determined, and the direction vector pointing from the end point to the inspection end point can be calculated through the coordinate difference. Then, according to the actual requirements, accuracy requirements, and spatial complexity, equipment distribution, etc., an appropriate extension step size is set. Subsequently, starting from the first initial trajectory end point, the extended points are generated sequentially along the direction vector according to the step size, and their coordinates are calculated through a specific formula. During the generation process, the new points need to be verified and adjusted. On the one hand, it is checked whether they are within the valid space range and free nodes of the octree map. If not, the step size or direction is adjusted. On the other hand, in combination with the actual requirements of the inspection for the perspective, distance, etc. of the equipment, it is ensured that the extended points can meet the inspection task.

[0082] Among them, the extended points can also be generated by the way of rapidly-exploring random tree (RRT). In the embodiments of the present application, in the direction from the end point on the first initial trajectory to the inspection end point, the extended points of the first initial trajectory are generated, including:

[0083] Generating random extended points in the direction from the end point on the first initial trajectory to the inspection end point;

[0084] Determining the extended points in the random extended points according to the positional relationship between the position of the random path point and the occupied node, and the positional relationship between the path from the random path point to the end point and the occupied node.

[0085] Among them, the random extended points can refer to a series of spatial position points generated randomly in the direction from the end point on the first initial trajectory to the inspection end point starting from the end point on the first initial trajectory. When these points are generated, instead of following a fixed rule, their specific positions are randomly determined within a certain range to explore various possible extended paths.

[0086] In order to find the optimal points, the random path points can be screened and judged according to the positional relationship between the position of the random path point and the occupied node, and the positional relationship between the path from the random path point to the end point and the occupied node, so as to determine the extended points that meet the requirements and can be actually used to optimize the inspection path.

[0087] Among them, the positional relationship between the position of the random path point and the occupied node can include:

[0088] Containment relationship: That is, it means that the random path point is completely inside the occupied node, indicating that the space area where the point is located is occupied by an object or obstacle. In the actual inspection path planning, this point is the position where there is a collision problem because the inspection equipment cannot pass through the obstacle.

[0089] Intersection relationship: That is, it means that the random path point may intersect with the boundary of the occupied node. It means that part of this point is inside the occupied node and part is outside the occupied node. In the actual inspection path planning, this point is the position where there is a collision risk because the inspection equipment may collide with obstacles.

[0090] Separation relationship: That is, it means that the random path point maintains a certain distance from all occupied nodes and is completely located in the free space. In the actual inspection path planning, this point is the position of a usable point.

[0091] The positional relationship between the path between the random path point and the end point and the occupied node can include:

[0092] The path has no intersection with the occupied node: The connection line or trajectory between the random path point and the end point is completely in the free space and does not intersect or touch any occupied node. This indicates that this path is completely feasible and is the preferred route.

[0093] The path has an intersection with the occupied node: The path partially passes through the occupied node, that is, during the process from the random path point to the end point, part of it enters the interior of the occupied node. This indicates that this path may have risks and is not recommended to be selected.

[0094] The path is tangent to or intersects with the boundary of the occupied node: The path has contact with the boundary of the occupied node, which may be tangent, that is, the path just grazes the boundary of the occupied node; or it may be intersecting, that is, the path has one or more intersection points with the boundary of the occupied node. This indicates that this path may have risks and is not recommended to be selected.

[0095] The path is completely contained in the occupied node: The path between the random path point and the end point is completely located inside one or more occupied nodes. This indicates that this path is completely infeasible and this path must be discarded in the path planning and other feasible path schemes must be searched again because the inspection equipment cannot pass through the space occupied by obstacles.

[0096] Therefore, according to the positional relationship between the position of the random path point and the occupied node, and the positional relationship between the path between the random path point and the end point and the occupied node, a random extension point that can drive safely and has the shortest line can be selected as the extension point.

[0097] S203. Adjust the first initial trajectory according to the shortest path between the extension point and other points on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map to obtain the first trajectory.

[0098] Among them, other points on the first initial trajectory may refer to points other than the end point, and this point may be the extension point when generating the first initial trajectory.

[0099] The shortest path between the extended point and other points on the first initial trajectory may refer to the straight-line distance between the extended point and other points. The path distance of the shortest path may refer to the length of the straight line, and the positional relationship between the shortest path and the occupied nodes in the octree map may refer to the positional relationship between the shortest path and the occupied nodes. This positional relationship is of the same type as the positional relationship between the path between the random path point and the end point and the occupied nodes, so it will not be elaborated here.

[0100] Adjusting the first initial trajectory may refer to adjusting the path connecting the extended point and the end point to connect the extended point and some other point. For example, the first initial trajectory is composed of four points 1, 2, 3, and 4, where 1, 2, and 3 are other points and 4 is the end point. When the shortest path between the extended point 5 and the other point 3 is less than the shortest path between the end point 4 and the extended point 5, and the positional relationship between the shortest path between the extended point 5 and the other point 3 and the occupied nodes in the octree map is that the path has no intersection with the occupied nodes, then the end point 4 is deleted, and the extended point 5 is connected to the other point 3. Thus, the first trajectory is obtained, which is composed of four points 1, 2, 3, and 5.

[0101] In the embodiment of the present application, since the straight line is the shortest between two points, after determining the extended point, the shortest path between the extended point and other points on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map are determined to adjust the first initial trajectory, so as to obtain the path with the shortest moving distance again.

[0102] Among them, in the embodiment of the present application, according to the shortest path between the extended point and other points on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map, the first initial trajectory is adjusted to obtain the first trajectory, including:

[0103] Determine the shortest path from the extended point to other points on the first initial trajectory respectively, and the path distance of the shortest path;

[0104] According to the positional relationship between the shortest path and the occupied nodes in the octree map, determine the target other point among the other points, the shortest path distance between the target other point and the extended point, and the length from the inspection start point along the first initial trajectory to the target other point;

[0105] According to the target other point among the other points, the shortest path distance between the target other point and the extended point, and the length from the inspection start point along the first initial trajectory to the target other point, adjust the first initial trajectory to obtain the first trajectory.

[0106] Among them, when determining the first trajectory, the first initial trajectory can be adjusted in sequence according to the distances between other points and the end point. For example, the first initial trajectory is composed of four points 1, 2, 3, and 4, where 1, 2, and 3 are other points and 4 is the end point. When the shortest path between the extended point 5 and the other point 3 is less than the shortest path between the end point 4 and the extended point 5, and the position relationship between the shortest path between the extended point 5 and the other point 3 and the occupied nodes in the octree map is that the path has no intersection with the occupied nodes, then the end point 4 is deleted. At this time, it is judged whether the shortest path between the extended point 5 and the other point 2 is less than the shortest path between the end point 3 and the extended point 5, and the position relationship between the shortest path between the extended point 5 and the other point 2 and the occupied nodes in the octree map. When the shortest path between the extended point 5 and the other point 2 is less than the shortest path between the end point 3 and the extended point 5, and the position relationship between the shortest path between the extended point 5 and the other point 2 and the occupied nodes in the octree map is that the path has no intersection with the occupied nodes, then the end point 3 is deleted, and the extended point 5 is connected to the other point 2. Thus, the first trajectory is obtained, which is composed of four points 1, 2, and 5.

[0107] Among them, in the embodiment of the present application, the angle between any two adjacent path segments on the first trajectory is greater than or equal to a preset angle threshold.

[0108] Among them, any two adjacent path segments on the first trajectory can refer to the path segments between three consecutive points. For example, when the first trajectory includes four points 1, 2, 3, and 4, then the three consecutive points can refer to the points 1, 2, 3, the points 2, 3, 4, and the points 2, 3, 4. The two adjacent path segments can refer to the path segment between point 1 and point 2 and the path segment between point 2 and point 3, and the path segment between point 2 and point 3 and the path segment between point 3 and point 4.

[0109] The preset angle threshold can determine the smoothness of the path. When the preset angle threshold is large, it means that the path is required to be smoother to avoid overly sharp turns. For example, in the path planning of an inspection UAV, if the preset angle threshold is 120°, then during the flight of the UAV, the angle between two adjacent flight paths must be greater than or equal to 120°. This can prevent the UAV from making frequent sharp turns, ensure the stability and safety of the flight, and also help reduce energy consumption.

[0110] S204. Generate an inspection trajectory according to the distance between the end point of the first trajectory and the end point of the second trajectory on the first trajectory. The second trajectory is the trajectory extending from the inspection end point to the inspection start point direction in the octree map.

[0111] Among them, in the spatial environment constructed by the octree map, in order to efficiently generate the inspection trajectory, the trajectory planning can be advanced simultaneously from two directions, namely the inspection starting point and the inspection ending point, that is, the first trajectory and the second trajectory.

[0112] Figure 2a This is a schematic diagram of generating the inspection trajectory provided by the embodiment of the present application. As Figure 2a shown, the generation process of the inspection trajectory includes:

[0113] First, generate the first trajectory starting from the inspection starting point. During the generation process of this trajectory, by determining the first initial trajectory, generating extension points based on the end point positions, and combining with the position relationship with the occupied nodes, etc., continuously optimize and adjust to ensure that the first trajectory covers the inspection area as efficiently as possible. At the same time, starting from the inspection ending point, extend and generate the second trajectory in the direction of the inspection starting point. During the generation process of these two trajectories, the spatial information in the octree map is fully considered, including the occupancy of nodes, the spatial relationship between each point, etc.

[0114] Then, calculate the distance between the end point of the first trajectory and the end point of the second trajectory. When this distance is less than a preset certain value, it indicates that the ends of the two trajectories are close enough. At this time, using the path search algorithm, in the space defined by the octree map, on the premise of avoiding obstacles (i.e., the occupied node area), search for an optimal connection path between the two end points. Combine this connection path with the first trajectory and the second trajectory to form a complete, efficient and practically space-constrained inspection trajectory.

[0115] In the embodiment of the present application, let G be a mapping composed of vertices V and edges E in the configuration space, and denote the straight line connecting point q1 and point q2 as LINE(q1, q2). The random exploration map G1 starts from the initial node qinit, which is also the starting position of the aircraft. Another random exploration map G2 is also initialized with the target node qgoal, which is located at the target position. The generation of qrand is selected by Sample(qinit, qgoal, p, G1). First, give heuristic information to generate a random tree in the direction of the qgoal point; then, check whether θn composed of LINE(qparent, qnearst) and LINE(qnearst, qrand) is suitable for flight. If θn is small, the UAV needs to fly a sharp turn to follow the path, and this small angle is not conducive to the normal flight of the UAV. Therefore, qrand with θn > 90 can be taken to ensure that the UAV avoids emergency braking, and perform local trimming on the mapping G1 through Reconnect(qnew, qnearest, G1). Then, the qnew point and the trimmed mapping G1 can be obtained. Among them, the algorithm for executing the above process can be:

[0116] 1: V1 ← {qinit}; E1’ ← ∅; G1 ← (V1, E1)

[0117] 2: V2 ← {qgoal}; E2’ ← ∅; G2 ← (V1, E1)

[0118] 3: while i < N do

[0119] 4: qrand ← Sample(qinit, qgoal, p, G1); i ← i + 1;

[0120] 5: if ObstacleFree(qnearest, qnew) then

[0121] 6: V1 ← V1 ∪ {qnew};

[0122] 7: E1 ← E1 ∪ {qnearest, qnew};

[0123] 8: q’nearest ← Nearest(G2, qnew);

[0124] 9: q’new ← Steer(G2, qnew);

[0125] 10: if ObstacleFree(q’nearest, q’new) then

[0126] 11: V2 ← V2 ∪ {q’new};

[0127] 12: E2 ← E2 ∪ {q’nearest, q’new};

[0128] 13: while not q’new = qnew

[0129] 14: q’’new ← Steer(q’new, qnew);

[0130] 15: if ObstacleFree(q’’new, q’new) then

[0131] 16: V2 ← V2 ∪ {q’’new};

[0132] 17: E2 ← E2 ∪ {q’’new, q’new};

[0133] 18: q’new ← q’’new;

[0134] 19: end if

[0135] 20: else break;

[0136] 21: end while

[0137] 22: end if

[0138] 23: end if

[0139] 24: end while

[0140] 25: if q’new=q’’new then

[0141] 26: return Reached;

[0142] 27: end if

[0143] 28: if abs (|V2|-|V1|)>Threshold then

[0144] 29: Swap (G2,G1)

[0145] 30: end if

[0146] The inspection path planning method provided by the embodiments of the present application first determines a first initial path extending from the inspection starting point towards the inspection ending point in the octree map, then generates extension points along the direction from the end point of the first initial path to the inspection ending point, and then adjusts the first initial path to obtain the first path based on the shortest path between the extension points and other points on the first initial path, the length of this shortest path, and the positional relationship between the shortest path and the occupied nodes representing obstacles in the map. Finally, a complete inspection path is generated according to the distance between the end point of the first path and the end point of the second path extending from the inspection ending point towards the inspection starting point. In this process, paths are planned simultaneously from two directions, which can quickly cover the inspection area and reduce the planning time. Generating extension points and adjusting the path in combination with information such as the shortest path can make the path more reasonably avoid obstacles. By judging that the distance between the end points of the two paths is less than a certain value and then connecting them, the two paths can be efficiently integrated. Thus, the effect of quickly, efficiently, and accurately generating an inspection path that meets the actual spatial constraints is achieved, the technical problem of how to quickly plan a comprehensive and reasonable inspection path in a complex spatial environment is solved, the inspection efficiency and accuracy are improved, and the inspection cost and risk are reduced.

[0147] Figure 3 is the flow diagram of the inspection path planning method provided by the present application Figure 2 ,as Figure 3 shown, this embodiment is in Figure 2Based on the embodiments, the process of generating an inspection trajectory according to the distance between the end point of the first trajectory and the end point of the second trajectory on the second trajectory is described in detail. The method includes:

[0148] S301. If the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory is less than a preset distance, an initial inspection trajectory is generated according to the first inspection trajectory and the second inspection trajectory;

[0149] S302. Smooth the connection points of the path segments in the initial inspection trajectory to obtain a smooth inspection trajectory;

[0150] S303. Optimize the smooth inspection trajectory according to the preset inspection speed condition and inspection acceleration condition to determine the inspection trajectory;

[0151] S304. If the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory is greater than or equal to the preset distance, the first trajectory is used as the first initial trajectory, and the step of determining the first initial trajectory in the octree map is re-executed until the distance between the end point of the first trajectory and the end point of the second trajectory is less than the preset distance, and an inspection trajectory is generated.

[0152] Among them, in the process of generating the inspection trajectory, first, the relationship between the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory and the preset distance is judged. If the distance is less than the preset distance, an initial inspection trajectory is generated according to the first inspection trajectory and the second inspection trajectory. Then, the connection points of the path segments in the initial inspection trajectory are smoothed to obtain a smooth inspection trajectory. Then, according to the preset inspection speed condition and inspection acceleration condition, the smooth inspection trajectory is optimized to finally determine the inspection trajectory; if the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory is greater than or equal to the preset distance, the current first trajectory is used as the new first initial trajectory, and the step of determining the first initial trajectory in the octree map is re-executed. In this way, the loop is repeated until the distance between the end point of the first trajectory and the end point of the second trajectory is less than the preset distance, and then an inspection trajectory is generated.

[0153] In the embodiments of the present application, optimizing the smooth inspection trajectory according to the preset inspection speed condition and inspection acceleration condition may be:

[0154] The inspection trajectory can be represented by an nth-order polynomial:

[0155] ;

[0156] Among them, is a trajectory parameter. Assume , the trajectory can be written in vector form: where n represents the order of the polynomial, i is the index, and t is the time. The trajectory is represented by a polynomial.

[0157] The snap of the trajectory can be calculated by the following formula:

[0158] ;

[0159] Among them, snap is an important optimization goal and is the rate of change of the acceleration of the UAV. By minimizing snap, the vibration and energy consumption during flight can be reduced, thereby improving the overall performance and flight quality of the aircraft. In the embodiments of the present application, snap can be calculated by taking the fourth derivative of the trajectory function.

[0160] Since the polynomial curve is too simple to represent complex trajectories, the trajectory is divided into multiple segments according to time, and each segment is represented by a polynomial curve. The purpose of trajectory planning is to find the polynomial parameters. Thus, the route can be optimized by minimizing snap, and the constructed optimization function is as follows:

[0161] ;

[0162] Among them, P is the vector composed of polynomial coefficients, and Q is the weight matrix.

[0163] At the same time, the positions of the intermediate path points, as well as the positions, velocities, and accelerations of the starting point and the ending point, can be used as equality constraint equations.

[0164] In the embodiments of the present application, the safety distance of power inspection can also be fused as a buffer to achieve trajectory optimization. Among them, the optimization formula can be:

[0165] ;

[0166] Among them, this formula represents that the generated intermediate trajectory points are in the bounding box of the starting point and the ending point, r represents the size of the bounding box, ti represents time, represents the position of the object at the time point ti.

[0167] Thus, Figure 3a is a schematic diagram of the optimization result provided by the present application. As shown in the optimization result of Figure 3a , the optimized inspection trajectory is smooth and convenient to meet the flight requirements of the UAV. Among them, for time allocation, the segment time can be minimized by respecting the state constraints of the maximum speed and acceleration.

[0168] The inspection trajectory planning method provided by the embodiments of the present application divides the flight path and obtains a smooth minimum bounce trajectory with buffer constraints, and optimizes the path into a safe, smooth, and dynamically feasible trajectory in real time, so as to realize the three-dimensional space inspection path planning in the power scenario, greatly improving the intelligence and automation level of power operation and maintenance.

[0169] Figure 4 It is a schematic structural diagram of the inspection trajectory planning device provided by the present application, as Figure 4 shown, the inspection trajectory planning device 40 provided in this embodiment includes:

[0170] A determination module 401, configured to determine a first initial trajectory in the octree map, where the first initial trajectory is a trajectory extending from the inspection start point to the inspection end point in the octree map;

[0171] A point generation module 402, configured to generate extended points of the first initial trajectory in the direction from the end point on the first initial trajectory to the inspection end point;

[0172] An adjustment module 403, configured to adjust the first initial trajectory according to the shortest path between the extended points and other points on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map, to obtain a first trajectory;

[0173] A trajectory generation module 404, configured to generate an inspection trajectory according to the distance between the end point of the first trajectory and the end point of the second trajectory on the second trajectory, where the second trajectory is a trajectory extending from the inspection end point to the inspection start point in the octree map.

[0174] In a possible implementation manner, the octree map in the determination module 401 is constructed according to the point cloud data of the space where the object to be inspected is located;

[0175] The types of occupied nodes in the octree map are different, and the resolutions of the occupied nodes are different.

[0176] In a possible implementation manner, the point generation module 402 may also be specifically configured to:

[0177] Generate random extended points in the direction from the end point on the first initial trajectory to the inspection end point;

[0178] Determine the extended points in the random extended points according to the positional relationship between the position of the random path points and the occupied nodes, and the positional relationship between the path between the random path points and the end point and the occupied nodes.

[0179] In a possible implementation manner, the adjustment module 403 may also be specifically configured to:

[0180] Determine the shortest paths from the extension points to other points on the first initial trajectory, and the path distances of the shortest paths;

[0181] According to the positional relationship between the shortest paths and the occupied nodes in the octree map, determine the target other points among the other points, the shortest path distance between the target other points and the extension points, and the length from the inspection starting point along the first initial trajectory to the target other points;

[0182] Adjust the first initial trajectory according to the target other points among the other points, the shortest path distance between the target other points and the extension points, and the length from the inspection starting point along the first initial trajectory to the target other points, to obtain the first trajectory.

[0183] In a possible implementation manner, the angle between any two adjacent path segments on the first trajectory in the adjustment module 403 is greater than or equal to a preset angle threshold.

[0184] In a possible implementation manner, the trajectory generation module 404 can also be specifically used for:

[0185] If the distance between the end point of the first trajectory and the end point of the second trajectory on the second trajectory is less than a preset distance, generate an initial inspection trajectory according to the first inspection trajectory and the second inspection trajectory;

[0186] Perform smoothing processing on the connection points of the path segments in the initial inspection trajectory to generate an inspection trajectory.

[0187] In a possible implementation manner, the trajectory generation module 404 can also be specifically used for:

[0188] Perform smoothing processing on the connection points of the path segments in the initial inspection trajectory to obtain a smooth inspection trajectory;

[0189] Optimize the smooth inspection trajectory according to the preset inspection speed condition and inspection acceleration condition to determine the inspection trajectory.

[0190] In a possible implementation manner, the trajectory generation module 404 can also be specifically used for:

[0191] If the distance between the end point of the first trajectory and the end point of the second trajectory on the second trajectory is greater than or equal to a preset distance, use the first trajectory as the first initial trajectory, and re-execute the step of determining the first initial trajectory in the octree map until the distance between the end point of the first trajectory and the end point of the second trajectory on the second trajectory is less than a preset distance, and generate an inspection trajectory.

[0192] The inspection trajectory planning device provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here in this embodiment.

[0193] Figure 5 This is a schematic structural diagram of the patrol route planning device provided by this application. As Figure 5 shown, the patrol route planning device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. Among them, the processor 501, the memory 502, and the communication component 503 are connected through a bus 504.

[0194] In the specific implementation process, at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that at least one processor 501 executes the above-mentioned method.

[0195] For the specific implementation process of the processor 501, reference can be made to the above method embodiment, and its implementation principle and technical effect are similar, so they will not be elaborated here in this embodiment.

[0196] In the above embodiment, it should be understood that the processor may be a central processing unit (English: Central Processing Unit, abbreviated as: CPU), and may also be other general-purpose processors, digital signal processors (English: Digital Signal Processor, abbreviated as: DSP), application-specific integrated circuits (English: Application Specific Integrated Circuit, abbreviated as: ASIC), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.

[0197] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0198] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0199] The present application also provides a computer program product, including a computer program which, when executed by a processor, implements the above-mentioned method.

[0200] The present application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-mentioned method.

[0201] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk or an optical disk. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0202] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0203] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0204] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0205] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0206] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., various media that can store program codes.

[0207] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When this program is executed, it executes the steps including the above method embodiments; and the aforementioned storage medium includes: ROMs, RAMs, magnetic disks, or optical discs, etc., various media that can store program codes.

[0208] Finally, it should be noted that: After considering the specification and practicing the invention disclosed herein, those skilled in the art will easily think of other implementation manners of the present invention. The present invention is intended to cover any variations, uses, or adaptive changes of the present invention. These variations, uses, or adaptive changes follow the general principles of the present invention and include the common general knowledge or conventional technical means in the technical field not disclosed in the present invention. It is not limited to the exact structure already described and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A patrol path planning method, characterized in that, Including: Determine a first initial trajectory in the octree map, where the first initial trajectory is a trajectory extending from the inspection start point to the inspection end point in the octree map; Generate extension points of the first initial trajectory in the direction from the end point position on the first initial trajectory to the inspection end point; Adjust the first initial trajectory according to the shortest path between the extension points and other points on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map to obtain a first trajectory; Generate an inspection trajectory according to the distance between the first trajectory end point on the first trajectory and the second trajectory end point on the second trajectory, where the second trajectory is a trajectory extending from the inspection end point to the inspection start point in the octree map.

2. The method according to claim 1, wherein The octree map is constructed based on the point cloud data of the space where the object to be inspected is located; The types of occupied nodes in the octree map are different, and the resolutions of the occupied nodes are different.

3. The method according to claim 1, wherein The generating extension points of the first initial trajectory in the direction from the end point position on the first initial trajectory to the inspection end point includes: Generate random extension points in the direction from the end point position on the first initial trajectory to the inspection end point; Determine the extension points among the random extension points according to the positional relationship between the position of the random path points and the occupied nodes, and the positional relationship between the path between the random path points and the end point position and the occupied nodes.

4. The method according to claim 1, wherein The adjusting the first initial trajectory according to the shortest path between the extension points and other points on the first initial trajectory, the path distance of the shortest path, and the positional relationship between the shortest path and the occupied nodes in the octree map to obtain a first trajectory includes: Determine the shortest paths from the extension points to other points on the first initial trajectory respectively, and the path distances of the shortest paths; According to the positional relationship between the shortest path and the occupied nodes in the octree map, determine the target other points among the other points, the shortest path distance between the target other points and the extension points, and the length from the inspection start point along the first initial trajectory to the target other points; Adjust the first initial trajectory according to the target other points among the other points, the shortest path distance between the target other points and the extension points, and the length from the inspection start point along the first initial trajectory to the target other points to obtain a first trajectory.

5. The method according to claim 4, characterized in that, The included angle between any two adjacent path segments on the first trajectory is greater than or equal to a preset angle threshold.

6. The method according to claim 1, wherein The generating an inspection trajectory according to the distance between the first trajectory end point on the first trajectory and the second trajectory end point on the second trajectory includes: If the distance between the first trajectory end point on the first trajectory and the second trajectory end point on the second trajectory is less than a preset distance, generate an initial inspection trajectory according to the first inspection trajectory and the second inspection trajectory; Perform smoothing processing on the joints of the path segments in the initial inspection trajectory to generate the inspection trajectory.

7. The method according to claim 6, wherein Smoothing the joints of path segments in the initial inspection trajectory to generate the inspection trajectory includes: Smoothing the joints of path segments in the initial inspection trajectory to obtain a smoothed inspection trajectory; Optimizing the smoothed inspection trajectory according to preset inspection speed conditions and inspection acceleration conditions to determine the inspection trajectory.

8. The method according to claim 6, characterized in that, The method further includes: If the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory is greater than or equal to a preset distance, then taking the first trajectory as the first initial trajectory and re-executing the step of determining the first initial trajectory in the octree map until the distance between the end point of the first trajectory on the first trajectory and the end point of the second trajectory on the second trajectory is less than the preset distance to generate an inspection trajectory.

9. An inspection path planning device, characterized in that, Including: A memory and a processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory, so that the processor executes the method according to any one of claims 1-8.

10. A computer program product, characterized in that, Including a computer program which, when executed by a processor, implements the method according to any one of claims 1-8.

Images