A power line inspection method and system based on drone

Abstract

The present invention provides a power line inspection method and system based on unmanned aerial vehicles, which relates to the field of unmanned aerial vehicle technology and automatic inspection of power lines. The method comprises: obtaining basic data of the unmanned aerial vehicle power inspection route to construct a two-dimensional environment map, wherein the basic data includes the specific layout of the power line, topographic information and obstacle positions; performing path planning according to the two-dimensional environment map, and obtaining a two-dimensional power line path planning according to the cost and actual distance of each node in the path; establishing a three-dimensional topographic map according to the two-dimensional power line path planning; constructing a cost function for inspection according to the three-dimensional topographic map; and evaluating the costs of different inspection paths according to the cost function to obtain an evaluated inspection path. The power line inspection method based on unmanned aerial vehicles of the present invention realizes efficient, safe, low-cost and high-quality power line inspection through precise path planning, optimization and evaluation.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) technology and automatic inspection of electric power lines, and in particular to an unmanned aerial vehicle (UAV)-based power line inspection method and system. Background Art

[0002] Traditional power line inspection methods mainly rely on manual work, and inspectors need to go to the site in person to inspect various power equipment one by one. However, this method has many disadvantages.

[0003] First, manual inspections are inefficient, especially when faced with vast power line networks, where inspectors often need to spend a lot of time and energy to complete their tasks. Second, complex terrain and adverse weather conditions pose great challenges to manual inspections. In some difficult-to-access areas such as mountainous areas and swamps, inspectors may face great safety risks and may even be unable to reach certain key checkpoints. In severe weather such as rain, snow, and lightning, inspections are even more difficult to carry out, which not only affects the timeliness of inspections, but may also threaten the lives of inspectors. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a UAV-based power line inspection method and system. The UAV-based power line inspection method realizes efficient, safe, low-cost and high-quality power line inspection through precise path planning, optimization and evaluation.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a power line inspection method based on a drone is provided, the method comprising:

[0007] Obtain basic data of the UAV power inspection route to construct a two-dimensional environment map, wherein the basic data includes the specific layout of the power lines, topographic information, and obstacle locations;

[0008] According to the two-dimensional environment map, path planning is performed, and the cost and actual distance of each node in the path are used to obtain the two-dimensional power line path planning;

[0009] Build a three-dimensional terrain map based on the two-dimensional power line path planning;

[0010] According to the three-dimensional terrain map, the cost function of the inspection is constructed;

[0011] According to the cost function, the costs of different inspection paths are evaluated to obtain the evaluated inspection paths;

[0012] Determine the planned three-dimensional path based on the evaluated inspection path;

[0013] The planned three-dimensional path is optimized to obtain an optimized three-dimensional path.

[0014] Furthermore, path planning is performed based on the two-dimensional environment map, and the cost and actual distance of each node in the path are used to obtain a two-dimensional power line path planning, including:

[0015] According to the two-dimensional environment map, mark the nodes, which include the starting point, end point, corner point, terrain feature point and point on the obstacle boundary of the power line;

[0016] Add the starting point of the node to the open list, where the list is a priority queue sorted by cost;

[0017] Select the node corresponding to the cost from the open list as the current node, remove the current node from the open list, add it to the closed list, and record the list of searched nodes. For each adjacent node of the current node: if the adjacent node includes the end point, the path planning is completed, and the complete path is obtained by backtracking; otherwise, calculate the cost of the adjacent node; if the adjacent node is not in the open list, update the cost of the adjacent node and set its parent node as the current node; if the adjacent node is not in the closed list, add it to the open list;

[0018] If the end point is added to the closed list, the path is found; starting from the end point, trace back to the starting point through the parent node relationship to obtain the two-dimensional power line path planning.

[0019] Furthermore, a three-dimensional terrain map is established based on the two-dimensional power line path planning, including:

[0020] Processing the two-dimensional power line terrain data to extract two-dimensional data, which includes key terrain feature points, elevation information, and obstacle locations;

[0021] Using two-dimensional data, a three-dimensional grid is generated, and each grid point is assigned a height value on the Z axis according to the corresponding elevation information to form a three-dimensional terrain surface;

[0022] Determine a corresponding texture image from a texture library according to the terrain type, so that the texture image is mapped onto the corresponding three-dimensional grid, wherein the terrain type includes grass, sand and rock;

[0023] Identify the locations of obstacles marked in the two-dimensional data and map them to corresponding locations in the three-dimensional terrain map;

[0024] The real lighting environment is simulated according to the preset lighting parameters, and corresponding light is generated on the terrain and obstacle surfaces through lighting calculation to obtain a three-dimensional terrain map, where the lighting parameters include the light source position and light intensity.

[0025] Furthermore, the two-dimensional data is used to generate a three-dimensional grid, and each grid point is assigned a height value on the Z axis according to the corresponding elevation information to form a three-dimensional terrain surface, including:

[0026] Create a grid structure and define a two-dimensional array to store the elevation values ​​of the grid points. Each element of the two-dimensional array corresponds to a grid point.

[0027] Traverse each point in the two-dimensional data and find the corresponding position in the three-dimensional grid based on the X-axis and Y-axis coordinates of each point;

[0028] Assign the elevation information of each point to the corresponding grid point in the grid structure;

[0029] Traverse each point in the grid structure, and generate a three-dimensional coordinate point based on the X-axis, Y-axis coordinates and elevation values ​​of each point;

[0030] A three-dimensional terrain surface is formed according to the three-dimensional coordinate points.

[0031] Furthermore, based on the three-dimensional terrain map, a cost function for inspection is constructed, including:

[0032] According to the characteristics of the UAV, obtain the relevant constraints of the UAV, including the maximum flight speed, maximum climb rate, maximum descent rate, turning radius and the minimum safe distance between the UAV and obstacles;

[0033] According to the three-dimensional terrain map and related constraints, the cost function of the inspection is constructed.

[0034] Furthermore, according to the cost function, the costs of different inspection paths are evaluated to obtain the evaluated inspection paths, including:

[0035] For each planned inspection path, the total cost of different inspection paths is calculated through the cost function;

[0036] Insert each path and its corresponding total cost as an element into the priority queue so that the queue automatically sorts the paths according to the total cost to obtain a priority queue;

[0037] The corresponding element is taken out from the priority queue as the evaluated inspection path.

[0038] Furthermore, according to the evaluated inspection path, a planned three-dimensional path is determined, including:

[0039] According to the evaluated inspection path, the number of inspection path points is determined;

[0040] Randomly initialize the position of each inspection path point in three-dimensional space;

[0041] In three-dimensional space, a plane is randomly selected as a mirror surface. For each inspection path point, the mirror position of the selected mirror surface is calculated to obtain the mirror position corresponding to the path point.

[0042] Compare the location of each inspection path point with the cost of its mirror image location to obtain a comparison result;

[0043] According to the comparison results, the planned three-dimensional path is determined.

[0044] Further, the planned three-dimensional path is optimized to obtain an optimized three-dimensional path, including:

[0045] Set the number of particles, each particle represents a three-dimensional path;

[0046] The position of each particle is randomly initialized in three-dimensional space. The position is represented as a series of three-dimensional coordinate points, forming a path.

[0047] Calculate the total cost of the path represented by each particle and use it as the fitness value of the particle;

[0048] Update the particle velocity and position until the preset number of iterations is reached to obtain the optimized three-dimensional path.

[0049] In the second aspect, a power line inspection system based on a drone includes:

[0050] The acquisition module is used to acquire the basic data of the UAV power inspection route to construct a two-dimensional environment map, wherein the basic data includes the specific layout of the power line, topographic information and the location of obstacles; according to the two-dimensional environment map, the path is planned, and the two-dimensional power line path planning is obtained according to the cost and actual distance of each node in the path; according to the two-dimensional power line path planning, a three-dimensional terrain map is established;

[0051] The processing module is used to construct a cost function for inspection based on a three-dimensional terrain map; evaluate the costs of different inspection paths based on the cost function to obtain an evaluated inspection path; determine a planned three-dimensional path based on the evaluated inspection path; and optimize the planned three-dimensional path to obtain an optimized three-dimensional path.

[0052] According to a third aspect, a computing device includes:

[0053] one or more processors;

[0054] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0055] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0056] The above solution of the present invention includes at least the following beneficial effects:

[0057] Through precise path planning and optimization, the UAV can fly along the shortest and safest path, thereby reducing unnecessary flight time and energy consumption. The terrain, obstacle locations and UAV flight capabilities are fully considered when planning the path, ensuring that the UAV can avoid obstacles during flight, reducing the risk of collision and enhancing the safety of inspections. By constructing a cost function and evaluating the costs of different inspection paths, the path with the lowest cost can be selected for inspection, thereby reducing the overall cost of inspections. The optimized three-dimensional path can ensure that the UAV flies in a more stable and reliable manner, reducing jitter and deviation during flight, thereby improving the accuracy and quality of inspection data. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a flow chart of a power line inspection method based on a drone provided in an embodiment of the present invention.

[0059] Figure 2 It is a schematic diagram of a power line inspection system based on a drone provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in a form and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0061] like Figure 1 As shown, an embodiment of the present invention provides a power line inspection method based on a drone, and the method comprises the following steps:

[0062] Step 11, obtaining basic data of the UAV power inspection route to construct a two-dimensional environment map, wherein the basic data includes the specific layout of the power lines, topographic information, and obstacle locations;

[0063] Step 12, performing path planning according to the two-dimensional environment map, and obtaining a two-dimensional power line path planning according to the cost and actual distance of each node in the path;

[0064] Step 13, establishing a three-dimensional terrain map based on the two-dimensional power line path planning;

[0065] Step 14, constructing a cost function for inspection based on the three-dimensional terrain map;

[0066] Step 15, evaluating the costs of different inspection paths according to the cost function to obtain an evaluated inspection path;

[0067] Step 16, determining a planned three-dimensional path according to the evaluated inspection path;

[0068] Step 17: Optimize the planned three-dimensional path to obtain an optimized three-dimensional path.

[0069] In an embodiment of the present invention, by obtaining the basic data of the UAV power inspection route, the environmental characteristics of the inspection area can be fully understood, and path planning of the two-dimensional environmental map can be performed, so that the optimal path connecting each key node can be efficiently found, and the cost of the node and the actual distance can be combined to make the path planning closer to the actual inspection needs; establishing a three-dimensional terrain map can more intuitively show the terrain changes in the inspection area, which is helpful to more accurately plan the flight trajectory of the UAV; by constructing a cost function for the inspection, the economic and time costs of different paths can be quantified, and the costs of different inspection paths can be evaluated according to the cost function, which is helpful to select the inspection path with the lowest cost and the highest efficiency, and improve the economic benefits of the inspection work; determining the planned three-dimensional path can ensure that the UAV can fly safely and accurately during the inspection process, avoiding collisions with obstacles or deviations from the planned route; optimizing the planned three-dimensional path can further reduce the redundant parts in the path and improve the flight efficiency and inspection speed of the UAV; evaluating and adjusting the optimized three-dimensional path can ensure the safety and feasibility of the final inspection path.

[0070] Step 11, obtaining basic data of the UAV power inspection route to construct a two-dimensional environment map, wherein the basic data includes the specific layout of the power lines, topographic information, and obstacle locations, including:

[0071] According to the needs of power inspection, the basic data types to be obtained are clearly defined, including the specific layout of power lines, topographic information, and obstacle locations;

[0072] Based on the basic data types, a two-dimensional environmental map is constructed through GIS.

[0073] In the embodiment of the present invention, by constructing a two-dimensional environment map, the layout of the power lines, the topography and the specific location of obstacles can be intuitively understood. It is crucial to clarify the location of obstacles for drone inspections. The two-dimensional environment map can provide a clear distribution of obstacles, avoid potential risks in advance, and prevent drones from colliding with obstacles. Through the two-dimensional environment map, the complexity and characteristics of the inspection area can be understood.

[0074] In a specific embodiment of the present invention, according to the needs of power inspection, basic data is obtained, and the data mainly includes the specific layout of power lines, topographic information and obstacle locations, wherein the power line layout data includes the direction of the line, the location of the tower, and the type of conductor; the topographic information covers the elevation change, terrain features and surface cover type of the ground; the obstacle location data records the location and size of buildings, trees or other structures around the line that may affect the inspection. The existing geographic information system (GIS) platform is used to conduct field surveys and collect data. The various types of data collected are integrated and processed to ensure the accuracy and consistency of the data. The processed power line layout, topographic and obstacle location data are imported into the GIS system, and the display properties and styles of the layers are adjusted, such as the color and width of the line, the contour lines or color bands of the terrain, and the symbolization of the obstacles, and spatial analysis is performed, such as buffer analysis and overlay analysis, to construct a two-dimensional environment map.

[0075] In a preferred embodiment of the present invention, the above step 12 performs path planning according to the two-dimensional environment map, and obtains a two-dimensional power line path planning according to the cost and actual distance of each node in the path, including:

[0076] According to the two-dimensional environment map, mark the nodes, which include the starting point, end point, corner point, terrain feature point and point on the obstacle boundary of the power line;

[0077] Add the starting point of the node to the open list, where the list is a priority queue sorted by cost;

[0078] Select the node corresponding to the cost from the open list as the current node, remove the current node from the open list, add it to the closed list, and record the list of searched nodes. For each adjacent node of the current node: if the adjacent node includes the end point, the path planning is completed, and the complete path is obtained by backtracking; otherwise, calculate the cost of the adjacent node; if the adjacent node is not in the open list, update the cost of the adjacent node and set its parent node as the current node; if the adjacent node is not in the closed list, add it to the open list;

[0079] If the end point is added to the closed list, the path is found; starting from the end point, trace back to the starting point through the parent node relationship to obtain the two-dimensional power line path planning.

[0080] In an embodiment of the present invention, the shortest or optimal path is quickly found through heuristic search. By accurately calculating the cost of each node and combining the actual distance and the heuristic estimated distance, an optimal or near-optimal path can be accurately found, which can adapt to different terrain and obstacle conditions. In the inspection of power lines, terrain feature points and points on obstacle boundaries are marked as nodes, and an inspection path that adapts to these conditions is generated. By recording the list of searched nodes and the parent node relationship, it is allowed to trace back to the starting point after finding the end point, so as to obtain a complete inspection path, which is convenient for adjusting or optimizing the path. Because the algorithm takes into account multiple factors, such as actual distance, terrain features, and obstacles; therefore, the generated path has strong robustness and can cope with the uncertainty and changes in the actual inspection process to a certain extent.

[0081] In a specific embodiment of the present invention, according to a two-dimensional map of the power line environment, which includes a starting point, an end point, a corner point, a terrain feature point and a point on the boundary of an obstacle, all important nodes, including the starting point, the end point and the corner point, are marked on the two-dimensional environment map. In specific implementation, the starting point is added to the open list: the starting point is taken as the initial current node and added to the open list, and the node with the smallest cost is selected from the open list as the current node, wherein the cost includes the actual distance from the starting point to the current node (or adjacent node) and the estimated distance from the current node (or adjacent node) to the end point; the current node is removed from the open list and added to the closed list; if so, the path planning is completed and the complete path can be obtained by backtracking. If the adjacent node includes an obstacle or is in the closed list, it is ignored and the actual cost of reaching the adjacent node from the starting point through the current node is calculated. If the current node is not in the open list, its cost is calculated and added to the open list, and its parent node is recorded as the current node. If the adjacent node is already in the open list, check whether a smaller cost can be obtained by including the current node. If so, update the cost of the adjacent node and the parent node as the current node. Repeat the above steps until the end point is added to the closed list or the open list is empty. When the end point is added to the closed list, it means that the path from the start point to the end point has been found. At this time, starting from the end point, tracing back to the start point through the parent node relationship of each node, the complete path can be obtained.

[0082] In a preferred embodiment of the present invention, the above step 13, establishing a three-dimensional terrain map based on the two-dimensional power line path planning, includes:

[0083] Processing the two-dimensional power line terrain data to extract two-dimensional data, which includes key terrain feature points, elevation information, and obstacle locations;

[0084] Using two-dimensional data, a three-dimensional grid is generated, and each grid point is assigned a height value on the Z axis according to the corresponding elevation information to form a three-dimensional terrain surface;

[0085] Determine a corresponding texture image from a texture library according to the terrain type, so that the texture image is mapped onto the corresponding three-dimensional grid, wherein the terrain type includes grass, sand and rock;

[0086] Identify the locations of obstacles marked in the two-dimensional data and map them to corresponding locations in the three-dimensional terrain map;

[0087] The real lighting environment is simulated according to the preset lighting parameters, and corresponding light is generated on the terrain and obstacle surfaces through lighting calculation to obtain a three-dimensional terrain map, where the lighting parameters include the light source position and light intensity.

[0088] In an embodiment of the present invention, a three-dimensional topographic map can intuitively display the relationship between the power line path and the surrounding terrain, and can clearly understand the line direction and terrain changes; by extracting key terrain feature points, elevation information and obstacle locations, the line layout and optimization can be accurately performed in three-dimensional space, avoiding visual misleading and errors that may exist in two-dimensional planning, and improving the accuracy of planning. Using a three-dimensional topographic map, the resources required for line construction, such as earthwork volume and material usage, can be more accurately evaluated, thereby achieving a reasonable allocation and effective use of resources and reducing construction costs. By accurately mapping the location of obstacles in a topographic map, potential safety risks can be identified in the planning stage, and corresponding avoidance measures can be taken to ensure the safe operation of power lines.

[0089] In a specific embodiment of the present invention, key terrain feature points, such as mountain tops, valley bottoms, and slope inflection points, are extracted through GIS software, and the elevation information of each feature point is extracted to form a data set containing X, Y coordinates and elevation (Z value), and the location of obstacles, such as trees and buildings, is identified and marked, and their two-dimensional coordinates are recorded. According to the scope and accuracy of the terrain data, the resolution and size of the three-dimensional grid are determined, and the entire terrain area is divided into regular grid units, each unit corresponds to a three-dimensional coordinate point, and each grid point is assigned an elevation value to ensure that each grid point has accurate X, Y, and Z coordinates to form a three-dimensional terrain surface. According to the terrain characteristics and elevation information, different terrain types, such as grass, sand, and rock, are identified, and a texture library is created for each terrain type, which contains corresponding texture images. For each grid unit, a suitable texture image is selected from the texture library according to its terrain type, and the selected texture image is applied to the corresponding three-dimensional grid using texture mapping technology to enhance the visual realism of the terrain. The marked obstacle two-dimensional coordinates are converted into three-dimensional coordinates, and the elevation of the obstacle is set to the terrain elevation of its location plus the height of the obstacle itself. Place the obstacle model at the corresponding position of the 3D terrain map to ensure that it is correctly aligned with the terrain surface, and simulate the real lighting environment, set the light source position and lighting intensity, and calculate the light distribution on the terrain and obstacle surface through the lighting model. Render according to the calculation results to generate a 3D terrain map with real lighting effects.

[0090] In a preferred embodiment of the present invention, the above step 13 generates a three-dimensional grid using two-dimensional data, and each grid point is assigned a height value on the Z axis according to the corresponding elevation information to form a three-dimensional terrain surface, including:

[0091] Create a grid structure and define a two-dimensional array to store the elevation values ​​of the grid points. Each element of the two-dimensional array corresponds to a grid point.

[0092] Traverse each point in the two-dimensional data and find the corresponding position in the three-dimensional grid based on the X-axis and Y-axis coordinates of each point;

[0093] Assign the elevation information of each point to the corresponding grid point in the grid structure;

[0094] Traverse each point in the grid structure, and generate a three-dimensional coordinate point based on the X-axis, Y-axis coordinates and elevation values ​​of each point;

[0095] A three-dimensional terrain surface is formed according to the three-dimensional coordinate points.

[0096] In an embodiment of the present invention, the three-dimensional grid generated by the two-dimensional data can accurately reflect the ups and downs and details of the actual terrain. Each grid point is assigned an accurate height value according to the corresponding elevation information, thereby forming a realistic terrain surface in three-dimensional space, which helps to simulate and analyze terrain features more realistically. The three-dimensional grid provides an intuitive way to visualize the elevation information in the two-dimensional data. By generating three-dimensional coordinate points and forming a three-dimensional terrain surface, it is easier to observe and understand the spatial distribution and change rules of the terrain, making complex terrain data easy to interpret. Spatial analysis on the three-dimensional grid can obtain more accurate results. Since each grid point has a clear three-dimensional coordinate, distance measurement, area calculation, slope analysis and other operations can be performed in three-dimensional space, providing an accurate data basis for the geographic information system (GIS). When formulating land planning, resource development, and environmental protection strategies, the three-dimensional terrain surface can be used for more comprehensive evaluation and analysis based on the three-dimensional terrain map, so as to make more scientific and reasonable decisions.

[0097] In a specific embodiment of the present invention, the range and resolution of the three-dimensional grid are determined, the range is usually determined by the boundary of the two-dimensional data, and the resolution is set according to the required level of detail. A two-dimensional array is created, the size of the two-dimensional array matches the resolution of the grid, and each array element corresponds to a point in the grid. Start traversing each point in the two-dimensional data set, which usually contains the coordinates of the X-axis and Y-axis and the corresponding elevation value. For each point in the two-dimensional data, find the corresponding position in the two-dimensional array according to its X-axis and Y-axis coordinates. Assign the elevation value of the two-dimensional data point to the corresponding element in the two-dimensional array, so that each element in the array now contains an elevation value, representing the height of the position on the Z axis. Traverse each element in the two-dimensional array, and the array already contains the elevation information of each grid point. For each element in the array, a three-dimensional coordinate point is generated according to the position (i.e., array index) and the elevation value. Specifically, the row index and column index of the array element can be converted into the coordinates of the X-axis and Y-axis respectively, and the value of the element (i.e., elevation) is used as the coordinate of the Z axis. These 3D coordinate points are stored in an appropriate data structure, and a 3D mesh is constructed from the generated 3D coordinate points by creating a series of triangles or polygons that connect adjacent points to form a continuous surface.

[0098] In a preferred embodiment of the present invention, the above step 14, constructing a cost function of the inspection according to the three-dimensional terrain map, includes:

[0099] According to the characteristics of the UAV, obtain the relevant constraints of the UAV, including the maximum flight speed, maximum climb rate, maximum descent rate, turning radius and the minimum safe distance between the UAV and obstacles;

[0100] According to the 3D terrain map and related constraints, Construct the cost function of the inspection, where D is the flight distance, T is the flight time, P is the power of the device; w1, w2, w3 and w4 are weight coefficients; Z i and are the elevation values ​​and average elevation values ​​associated with terrain roughness, N is the number of points considered in the three-dimensional terrain map when calculating terrain roughness, C is the cost function; i represents the index of the point considered in the three-dimensional terrain map.

[0101] In an embodiment of the present invention, by constructing a cost function, multiple factors such as flight distance, flight time, equipment power and terrain roughness can be comprehensively considered to find the optimal inspection path. The cost function includes constraints such as the minimum safe distance between the drone and obstacles, which can ensure that the drone always maintains a safe distance from obstacles during the inspection process, reduce the risk of collision, and ensure the safe completion of the inspection task. The power of the equipment is an important factor in the cost function. By optimizing the inspection path, the energy consumption of the drone can be reduced, thereby reducing the operating cost of the inspection. The cost function takes into account the elevation value and the average elevation value related to the terrain roughness, so that the drone can better adapt to different terrain environments.

[0102] In a preferred embodiment of the present invention, the above step 15, evaluating the costs of different inspection paths according to the cost function to obtain the evaluated inspection paths, includes:

[0103] For each planned inspection path, the total cost of different inspection paths is calculated through the cost function;

[0104] Insert each path and its corresponding total cost as an element into the priority queue so that the queue automatically sorts the paths according to the total cost to obtain a priority queue;

[0105] The corresponding element is taken out from the priority queue as the evaluated inspection path.

[0106] In an embodiment of the present invention, the total cost of different inspection paths is calculated through a cost function, so that the economic feasibility of each path can be clearly understood. By automatically sorting the paths according to the total cost using a priority queue, the inspection path with the best cost can be quickly identified. The evaluated inspection paths provide an important reference for enterprise decision makers. After understanding the costs of different inspection paths, resources such as manpower, material resources, and financial resources can be allocated more reasonably. The use of advanced cost evaluation methods and data structures (such as priority queues) can promote enterprise innovation in the field of inspection technology. By continuously optimizing the evaluation process and methods, more efficient and intelligent inspection solutions can be explored.

[0107] In a specific embodiment of the present invention, multiple possible inspection paths are generated, and the paths meet the basic flight constraints of the UAV, such as the maximum flight distance and the turning radius; ensure that each planned path avoids obstacles and maintains a minimum safe distance from obstacles. For each planned inspection path, its total cost is calculated by a defined cost function. Substitute each path into the cost function to obtain the total cost value of the path. Create a priority queue, and insert each path and its calculated total cost as an element into the priority queue. The priority of the queue is based on the total cost, and the lower the cost, the higher the priority of the path. Take out the highest priority element (i.e., the path with the lowest total cost) from the priority queue, where the highest priority element will be the optimal inspection path after evaluation.

[0108] In a preferred embodiment of the present invention, the above step 16, determining the planned three-dimensional path according to the evaluated inspection path, includes:

[0109] According to the evaluated inspection path, the number of inspection path points is determined;

[0110] Randomly initialize the position of each inspection path point in three-dimensional space;

[0111] In three-dimensional space, a plane is randomly selected as a mirror surface. For each inspection path point, the mirror position of the selected mirror surface is calculated to obtain the mirror position corresponding to the path point.

[0112] Compare the location of each inspection path point with the cost of its mirror image location to obtain a comparison result;

[0113] According to the comparison results, the planned three-dimensional path is determined.

[0114] In an embodiment of the present invention, by randomly initializing the position of each inspection path point in three-dimensional space, and combining iterative updates and mirror position comparisons, it is helpful to find a globally optimal or near-optimal inspection path. By adjusting the maximum number of iterations, the iteration speed, and the selection parameters of the mirror position, the path planning can be adjusted and optimized according to actual conditions. Due to the use of random initialization and iterative update strategies, it has strong robustness. Even if the selection of the initial path point is not optimal, it can gradually approach the optimal solution through iterative updates and mirror position comparisons. By recording and updating the inspection path with the lowest cost, and using mirror positions for comparison, a better inspection path can be found within a fewer number of iterations, thereby improving the efficiency of path planning and reducing computing time and resource consumption.

[0115] In a specific embodiment of the present invention, the number of inspection path points that need to be set is determined based on the evaluated inspection path, and these path points will serve as key navigation points for the drone during the inspection process. In three-dimensional space, the positions of these inspection path points are reasonably distributed and initialized according to the shape and length of the inspection path to ensure that the path points can cover the entire inspection path and take into account the flight characteristics and constraints of the drone. Set the maximum number of iterations, and in each iteration, use the cost function to calculate the total cost of the current inspection path and record it. Try to fine-tune the position of each inspection path point to find a path configuration with lower cost. If the total cost can be reduced by fine-tuning the position of a path point, update the position of the path point. Repeat this process until the maximum number of iterations is reached or the total cost is no longer significantly reduced.

[0116] In another specific embodiment of the present invention, a plane is randomly selected in three-dimensional space as a mirror, and this mirror will be used to calculate the mirror position of each inspection path point. For each inspection path point, its mirror position on the selected mirror is calculated according to the principle of geometric transformation (such as symmetry transformation). The cost of each inspection path point and its mirror position is calculated separately using a cost function. The two cost values ​​are compared to determine whether to update the path point to its mirror position. If the cost of the mirror position is lower, the position of the path point is updated to its mirror position; otherwise, the original position remains unchanged. Repeat the above steps of iterative optimization and mirror position comparison until the maximum number of iterations is reached or other stopping conditions are met. After multiple iterations, a planned three-dimensional inspection path is generated based on the final inspection path point position. This path will serve as a reference trajectory for the drone to perform inspection tasks.

[0117] In a preferred embodiment of the present invention, the above step 17, optimizing the planned three-dimensional path to obtain an optimized three-dimensional path, includes:

[0118] Set the number of particles, each particle represents a three-dimensional path;

[0119] The position of each particle is randomly initialized in three-dimensional space. The position is represented as a series of three-dimensional coordinate points, forming a path.

[0120] Calculate the total cost of the path represented by each particle and use it as the fitness value of the particle;

[0121] Update the particle velocity and position until the preset number of iterations is reached to obtain the optimized three-dimensional path.

[0122] In an embodiment of the present invention, through the particle swarm optimization algorithm, it is possible to search and optimize the path in three-dimensional space, so as to find a patrol path with lower total cost and higher efficiency. The optimized path not only reduces the flight distance and time, but also may reduce energy consumption and risk. The particle swarm optimization algorithm has a strong global search capability and can adaptively adjust the speed and position of particles to adapt to different terrains and obstacle distributions, so that the algorithm has higher flexibility and robustness when dealing with complex and changeable three-dimensional environments. The particle swarm optimization algorithm usually has a faster convergence speed and less computing resource requirements, and can obtain optimization results in a shorter time, thereby improving the efficiency of path planning.

[0123] In a specific embodiment of the present invention, the number of particles N is set, and each particle will represent a possible three-dimensional path. The position of each particle is randomly initialized in three-dimensional space. The position is a series of three-dimensional coordinate points. These points are connected to form a path. The position of each particle can be expressed as a list or array of coordinate points. The speed is randomly initialized for each particle. The speed is a list or array composed of the change amount of three-dimensional coordinate points, which determines the moving direction and step size of the particle in the search space. For each particle, the total cost of the path represented by it is calculated using the previous cost function, where the total cost will be used as the fitness value of the particle to evaluate the quality of the path, where the fitness value is equal to the inverse of the cost function. Record the individual optimal position and the global optimal position of each particle. The individual optimal position is the best position found by the particle during the search process (i.e., the position with the lowest cost), and the global optimal position is the best position found among all particles. According to the individual optimal position and the global optimal position, the particle speed and position are adjusted to adjust the speed and position of each particle. Ensure that the updated particle position still meets the flight constraints of the drone, such as the maximum flight speed, turning radius, etc. If not, the position needs to be appropriately adjusted. Until the preset number of iterations is reached.

[0124] like Figure 2 As shown, an embodiment of the present invention further provides a power line inspection system 20 based on a drone, comprising:

[0125] The acquisition module 21 is used to acquire basic data of the UAV power inspection route to construct a two-dimensional environment map, wherein the basic data includes the specific layout of the power line, topographic information, and obstacle locations; perform path planning based on the two-dimensional environment map, and obtain a two-dimensional power line path planning based on the cost and actual distance of each node in the path; and establish a three-dimensional terrain map based on the two-dimensional power line path planning;

[0126] The processing module 22 is used to construct a cost function for inspection based on the three-dimensional terrain map; evaluate the costs of different inspection paths based on the cost function to obtain an evaluated inspection path; determine a planned three-dimensional path based on the evaluated inspection path; and optimize the planned three-dimensional path to obtain an optimized three-dimensional path.

[0127] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0128] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0129] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0130] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A power line inspection method based on drone, characterized in that: The method comprises: Obtain basic data of the UAV power inspection route to construct a two-dimensional environment map, wherein the basic data includes the specific layout of the power lines, topographic information, and obstacle locations; According to the two-dimensional environment map, path planning is performed, and the cost and actual distance of each node in the path are used to obtain the two-dimensional power line path planning; Build a three-dimensional terrain map based on the two-dimensional power line path planning; According to the three-dimensional terrain map, the cost function of the inspection is constructed; According to the cost function, the costs of different inspection paths are evaluated to obtain the evaluated inspection paths; Determine the planned three-dimensional path based on the evaluated inspection path; Optimizing the planned three-dimensional path to obtain an optimized three-dimensional path; According to the two-dimensional environment map, path planning is performed, and the cost and actual distance of each node in the path are used to obtain the two-dimensional power line path planning, including: According to the two-dimensional environment map, mark the nodes, which include the starting point, end point, corner point, terrain feature point and point on the obstacle boundary of the power line; Add the starting point of the node to the open list, where the list is a priority queue sorted by cost; Select the node corresponding to the cost from the open list as the current node, remove the current node from the open list, add it to the closed list, and record the list of searched nodes. For each adjacent node of the current node: if the adjacent node includes the end point, the path planning is completed, and the complete path is obtained by backtracking; otherwise, calculate the cost of the adjacent node; if the adjacent node is not in the open list, update the cost of the adjacent node and set its parent node as the current node; if the adjacent node is not in the closed list, add it to the open list; If the end point is added to the closed list, the path is found; starting from the end point, trace back to the starting point through the parent node relationship to obtain the two-dimensional power line path planning; According to the 3D terrain map, the cost function of the inspection is constructed, including: According to the characteristics of the UAV, obtain the relevant constraints of the UAV, including the maximum flight speed, maximum climb rate, maximum descent rate, turning radius and the minimum safe distance between the UAV and obstacles; According to the 3D terrain map and related constraints, Construct the cost function of the inspection, where is the flight distance, is the flight time, is the power of the device; , , and is the weight coefficient; and is the elevation value and average elevation value related to terrain roughness, is the number of points considered in the 3D terrain map when calculating terrain roughness, is the cost function; represents the index of the point considered in the three-dimensional topographic map; According to the cost function, the costs of different inspection paths are evaluated to obtain the evaluated inspection paths, including: For each planned inspection path, the total cost of different inspection paths is calculated through the cost function; Insert each path and its corresponding total cost as an element into the priority queue so that the queue automatically sorts the paths according to the total cost to obtain a priority queue; Take the corresponding element from the priority queue as the evaluated inspection path; Based on the evaluated inspection path, determine the planned three-dimensional path, including: According to the evaluated inspection path, the number of inspection path points is determined; Randomly initialize the position of each inspection path point in three-dimensional space; In three-dimensional space, a plane is randomly selected as a mirror surface. For each inspection path point, the mirror position of the selected mirror surface is calculated to obtain the mirror position corresponding to the path point. Compare the location of each inspection path point with the cost of its mirror image location to obtain a comparison result; According to the comparison results, the planned three-dimensional path is determined.

2. The power line inspection method based on drone according to claim 1 is characterized in that: According to the two-dimensional power line path planning, a three-dimensional terrain map is established, including: Processing the two-dimensional power line terrain data to extract two-dimensional data, which includes key terrain feature points, elevation information, and obstacle locations; Using two-dimensional data, a three-dimensional grid is generated, and each grid point is assigned a height value on the Z axis according to the corresponding elevation information to form a three-dimensional terrain surface; Determine a corresponding texture image from a texture library according to the terrain type, so that the texture image is mapped onto the corresponding three-dimensional grid, wherein the terrain type includes grass, sand and rock; Identify the locations of obstacles marked in the two-dimensional data and map them to corresponding locations in the three-dimensional terrain map; The real lighting environment is simulated according to the preset lighting parameters, and corresponding light is generated on the terrain and obstacle surfaces through lighting calculation to obtain a three-dimensional terrain map, where the lighting parameters include the light source position and light intensity.

3. The power line inspection method based on drone according to claim 2 is characterized in that: Using two-dimensional data, a three-dimensional grid is generated. Each grid point is assigned a height value on the Z axis according to the corresponding elevation information to form a three-dimensional terrain surface, including: Create a grid structure and define a two-dimensional array to store the elevation values ​​of the grid points. Each element of the two-dimensional array corresponds to a grid point. Traverse each point in the two-dimensional data and find the corresponding position in the three-dimensional grid based on the X-axis and Y-axis coordinates of each point; Assign the elevation information of each point to the corresponding grid point in the grid structure; Traverse each point in the grid structure and generate a three-dimensional coordinate point based on the X-axis, Y-axis coordinates and elevation value of each point; A three-dimensional terrain surface is formed according to the three-dimensional coordinate points.

4. The power line inspection method based on drone according to claim 3 is characterized in that: The planned three-dimensional path is optimized to obtain an optimized three-dimensional path, including: Set the number of particles, each particle represents a three-dimensional path; The position of each particle is randomly initialized in three-dimensional space. The position is represented as a series of three-dimensional coordinate points, forming a path. Calculate the total cost of the path represented by each particle and use it as the fitness value of the particle; Update the particle velocity and position until the preset number of iterations is reached to obtain the optimized three-dimensional path.

5. A power line inspection system based on drones, characterized in that: Applied to the method according to any one of claims 1 to 4, comprising: The acquisition module is used to acquire the basic data of the UAV power inspection route to construct a two-dimensional environment map, wherein the basic data includes the specific layout of the power line, topographic information and the location of obstacles; according to the two-dimensional environment map, the path is planned, and the two-dimensional power line path planning is obtained according to the cost and actual distance of each node in the path; according to the two-dimensional power line path planning, a three-dimensional terrain map is established; The processing module is used to construct a cost function for inspection based on a three-dimensional terrain map; evaluate the costs of different inspection paths based on the cost function to obtain an evaluated inspection path; determine a planned three-dimensional path based on the evaluated inspection path; and optimize the planned three-dimensional path to obtain an optimized three-dimensional path.

6. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

Images