Path planning method for inspection of orbital robots for space steel structures

Abstract

This invention discloses a method for track robot inspection path planning for spatial steel structures, comprising the following steps: S1: Calculate the net height of the spatial steel structure for each grid in the three-dimensional space to obtain the passable three-dimensional spatial area, convert the three-dimensional space into a two-dimensional planar grid, and calculate the set of key point coordinates; S2: Based on the set of key point coordinates, perform grouped solutions to obtain the optimal path and the total path length; S3: Input the key point coordinates corresponding to the optimal path into the three-dimensional space to verify the rationality of the planned path. This invention solves the problem of track path layout in complex three-dimensional spaces of railway passenger station steel structures. Through the algorithm, it achieves the optimal path, realizing the optimal combination of shortest track length, minimum number of robots, and maximum inspection coverage, while saving robot power and extending travel time. It is widely used in spatial steel structures in industries such as railway passenger stations, airports, rail transit, and stadiums.

Description

Technical Field

[0001] This invention belongs to the technical field of track robot surveying and navigation, and in particular, a track robot inspection path planning method for space steel structures. Background Technology

[0002] As railway passenger station steel structures age, the main steel structure suffers from significant corrosion and damage due to atmospheric corrosion, train wind vibration, and roof leaks. Metal roofs also exhibit rust, windblowing, and leaks under extreme weather conditions such as heavy rain and typhoons, seriously affecting railway operational safety and passenger safety. Current inspection methods, primarily relying on manual visual inspection, cannot meet the management requirements for daily, periodic, and emergency inspections of railway passenger station steel structures. An automated inspection solution is needed to replace manual labor; therefore, a spatial steel structure track inspection robot was developed, with the robot primarily responsible for inspecting the steel structure.

[0003] Track-based inspection robots are well-established in industries such as power rooms and oil pipelines, and their inspection paths are generally quite simple. For example, pipeline inspections often follow linear paths, while power room inspections target the server racks. However, applying this simple path approach to spatial steel structures presents a challenge due to the lack of suitable path planning algorithms.

[0004] The inspection space for the steel structure of railway passenger stations is a three-dimensional space characterized by limited height, dense structural members, numerous obstacles such as pipelines, and a lack of clear patterns. The planar contours of the steel structures vary and are irregular across different stations. The inspection targets include steel structural members at different elevations, bolt connections, purlin components, hangers, and ceiling panels, which are diverse, numerous, complex in their relationships, and highly repetitive. Therefore, it is necessary to consider many factors such as the robot's inspection path, image acquisition coverage, travel height, endurance, climbing and turning capabilities.

[0005] Currently, the single-path approach is too fixed and cannot cover all key inspection areas, leading to the omission of certain parts or nodes, affecting the comprehensiveness and accuracy of the inspection. Furthermore, due to the confined space or irregular structure of the spatial steel structure, a single path may cause the robot to get stuck in a certain position or fail to reach the target location, making it impossible to dynamically adjust the path or avoid obstacles according to real-time inspection needs. Additionally, the robot inspection requires a 3D point cloud modeling environment beforehand, but this modeling method suffers from slow scanning speed and high resource consumption. Applying this to complex spaces like the spatial steel structures of railway passenger stations would severely impact the robot's inspection efficiency, and the robot's limited battery life would restrict its travel distance, reducing the comprehensiveness and efficiency of the inspection.

[0006] Therefore, a path planning method for tracked robot inspection of space steel structures is proposed. This method is based on the principle of multi-objective optimization, aiming to optimize the inspection path, improve the scanning and photography efficiency and coverage during the inspection process, and enhance the endurance of the inspection robot, so as to efficiently and comprehensively complete the inspection task of space steel structures. Summary of the Invention

[0007] In response to the problems mentioned above, this invention proposes a path planning method for tracked robot inspection of space steel structures, the specific solution of which is as follows:

[0008] A method for path planning of a tracked robot for inspecting space steel structures includes the following steps:

[0009] S1: Establish a three-dimensional solid model of the spatial steel structure, calculate the net height of the spatial steel structure, compare it with the minimum passage height of the inspection robot, select areas where the net height of the spatial steel structure is not less than the minimum passage height of the inspection robot, define them as the passable area range, confirm the outermost edge line of the passable area, and extract the planar coordinate values ​​of the upper chord node to form the planar coordinate set of the upper chord node.

[0010] To determine the minimum recognition distance of the inspection robot, the set of planar coordinates of the upper chord node is approximately divided into rows and columns. The average spacing between rows and columns is calculated to form a set of key point coordinates.

[0011] S2: Preset the starting point of the inspection path, solve the optimal path through the key point coordinate value TSP, obtain the total path length, divide the total path length by the longest driving distance of the inspection robot, and take the integer part to obtain the minimum number of inspection robots k required.

[0012] When k > 1, the planar coordinates of the upper chord nodes are divided into k+1 groups. The TSP within each group is solved for the coordinates of the key points in each group to obtain the optimal path for each group. The optimal total path length of the k+1 groups is then obtained by summing the results.

[0013] When k≤1, ignore the grouped TSP solution and directly output the TSP solution result to obtain the optimal path and the optimal total path length;

[0014] S3: Input the key point coordinates, starting point coordinates and optimal path of group k+1 into the 3D solid model, form the track layout path according to the minimum layout spacing of the inspection robot's hanging structure, and establish corresponding 3D solid models of the hanging structure, track and robot to verify the rationality of the path;

[0015] If no collision occurs and the minimum safe distance requirement is met, the planned path is considered reasonable, and the optimal path layout diagram is output.

[0016] If the planned path is unreasonable, return to step S2 and regroup to solve the problem.

[0017] Preferably, in step S1, when the average spacing between rows and columns is greater than twice the minimum recognition distance of the inspection robot, the planar coordinates of all upper chord nodes are used to form a key point coordinate set; when the average spacing between rows and columns is less than twice the minimum recognition distance of the inspection robot, a sampling method is used to extract key point coordinate values ​​from the planar coordinates of the upper chord nodes to form a key point coordinate set.

[0018] A further preferred sampling method is to retain the coordinates of the outermost edge line, establish a sliding window with a length and width of twice the minimum recognition distance, and slide the window in rows and columns starting from the coordinates of the outermost edge line. Each window retains only one point, which is called a key point.

[0019] Preferably, in step S1, the calculation of the net height of the spatial steel structure involves dividing the surfaces of the upper chord and lower chord in the established three-dimensional solid model of the spatial steel structure into planar meshes, calculating the coordinates of the center point of each mesh, connecting the center points of the upper chord mesh and the lower chord mesh, calculating the length of each connecting line, and obtaining the net height of the spatial steel structure.

[0020] Preferably, in step S1, the K-means clustering algorithm is used to collect and divide the plane coordinate set of the upper chord node into X-direction coordinates and Y-direction coordinates, and the average spacing of rows and columns is calculated.

[0021] Preferably, in step S2, the optimal path refers to the path that starts from the starting point of the inspection path, passes through each key point, and reaches the destination. It must satisfy the condition that the key points are passed only once and the total path length is the shortest.

[0022] Preferably, in step S2, the TSP solution includes any one or more combinations of the following algorithms: serpentine algorithm, greedy algorithm, Dijkstra's algorithm, genetic algorithm, or ant colony algorithm.

[0023] Another object of the present invention is to provide an inspection robot that applies the aforementioned path planning method for track robot inspection of space steel structures.

[0024] The beneficial effects of this invention are:

[0025] 1. An inspection path planning algorithm suitable for space steel structure track robots is proposed, which solves the problem of complex three-dimensional space track path layout in railway passenger station steel structures. The algorithm realizes the optimal path and achieves the optimal combination of shortest track length, minimum number of robots, and maximum inspection coverage.

[0026] 2. By using the optimal path and performing calculations before the inspection robot begins its inspection, the final optimized walking path is obtained. Subsequently, the robot can walk along the specified path, saving battery power and extending the walking time.

[0027] 3. The track path arranged by the algorithm can be used for triangular spatial steel structures of different structural types such as space frames and trusses. It is also applicable to spatial steel structures with different projected areas and spatial steel structures with irregular planar layouts. The algorithm has universality and versatility and can be widely used in spatial steel structures in industries such as railway passenger stations, airports, rail transit, and stadiums.

[0028] 4. The proposed inspection path planning algorithm achieves full coverage of inspection of the complex three-dimensional space of railway passenger station steel structure. The deployed inspection track robot can meet the needs of intelligent inspection of railway passenger station steel structure, replace the original manual inspection, realize timed and fixed-route inspection and real-time early warning, improve the operation and maintenance management level and decision-making efficiency of railway passenger station, and ensure the safe operation of railway passenger station. Attached Figure Description

[0029] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. However, it should be understood that these drawings are designed for illustrative purposes only and are not intended to limit the scope of the present invention. Furthermore, unless specifically indicated, these drawings are intended only to conceptually illustrate the structural construction described herein and are not necessarily drawn to scale.

[0030] Figure 1 This is a flowchart illustrating an application example of the present invention;

[0031] Figure 2 This is a schematic diagram of an application example in this invention. Detailed Implementation

[0032] First, it should be noted that the specific structure, features, and advantages of the present invention will be described in detail below by way of examples. However, all descriptions are for illustrative purposes only and should not be construed as limiting the present invention in any way. Furthermore, any single technical feature described or implied in the embodiments mentioned herein, or any single technical feature shown or implied in the accompanying drawings, can still be arbitrarily combined or deleted among these technical features to obtain more other embodiments of the present invention that may not be directly mentioned herein. Additionally, for the sake of simplifying the drawings, the same or similar technical features may be indicated only in one place in the same drawing.

[0033] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "setting," "connection," "fixing," "screw connection," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal connection of two components or the interaction between two components. Unless otherwise explicitly limited, those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0034] The following is in conjunction with the appendix Figure 1 -Appendix Figure 2 This invention will be described in detail.

[0035] Example 1:

[0036] A method for path planning of a tracked robot for inspecting space steel structures includes the following steps:

[0037] S1: Establish a three-dimensional solid model of the spatial steel structure, calculate the net height of the spatial steel structure, compare it with the minimum passage height of the inspection robot, select areas where the net height of the spatial steel structure is not less than the minimum passage height of the inspection robot, define them as the passable area range, confirm the outermost edge line of the passable area, and extract the planar coordinate values ​​of the upper chord node to form the planar coordinate set of the upper chord node.

[0038] To determine the minimum recognition distance of the inspection robot, the set of planar coordinates of the upper chord node is approximately divided into rows and columns. The average spacing between rows and columns is calculated to form a set of key point coordinates.

[0039] S2: Preset the starting point of the inspection path, solve the optimal path through the key point coordinate value TSP, obtain the total path length, divide the total path length by the longest driving distance of the inspection robot, and take the integer part to obtain the minimum number of inspection robots k required.

[0040] When k > 1, the set of key point coordinates is divided into k+1 groups. The TSP within each group is solved for the key point coordinates to obtain the optimal path for each group. The optimal total path length of the k+1 groups is then summed.

[0041] When k≤1, ignore the grouped TSP solution and directly output the TSP solution result to obtain the optimal path and the optimal total path length;

[0042] S3: Input the key point coordinates, starting point coordinates and optimal path of group k+1 into the 3D solid model, form the track layout path according to the minimum layout spacing of the inspection robot's hanging structure, and establish corresponding 3D solid models of the hanging structure, track and robot to verify the rationality of the path;

[0043] If no collision occurs and the minimum safe distance requirement is met, the planned path is considered reasonable, and the optimal path layout diagram is output.

[0044] If the planned path is unreasonable, return to step S2 and regroup to solve the problem.

[0045] Working principle:

[0046] This invention eliminates the need for inspection robots to scan a 3D space and calculate the optimal path on-site, as is done in existing technologies, which consume significant amounts of power and shorten battery life. Instead, this process is moved to the pre-processing stage. Before the inspection robot's track inspection, a 3D solid model of the track space to be inspected is created using relevant software. This model is then converted into a 2D planar mesh, forming a set of key point coordinates. TSP (Tracking Point Optimization) grouping and solving steps are then performed to obtain the minimum number of inspection robots, the optimal path, and the total path length. The key point coordinates corresponding to the optimal path are then input into the 3D space to verify the rationality of the planned path. Based on the verified final path, the layout scheme is further refined, enabling the inspection robot to patrol along the optimal and reasonable path, maximizing navigation capabilities, conserving battery power, and avoiding obstacles to complete the inspection task.

[0047] In this embodiment, in step S1, a three-dimensional solid model of the spatial steel structure is established using BIM software such as Rhino and Revit. The three-dimensional solid model includes the main load-bearing components of the spatial steel structure, such as the upper chord, web members, and lower chord, as well as the skin plate components formed on the surfaces of the upper and lower chords. The minimum passage height of the inspection robot is determined based on the robot selection and hanging position, including the height of the robot body, the height of the boom, and the minimum safety distance. The minimum recognition distance of the inspection robot is determined comprehensively based on the acquisition capabilities of the selected robot's visible light camera, thermal infrared camera, lidar, and other image acquisition equipment, as well as the ability to identify the defects to be identified.

[0048] Based on the three-dimensional solid model, the minimum passage height of the inspection robot, and the minimum recognition distance of the inspection robot obtained above, the set of key point coordinates is obtained.

[0049] In step S2, the TSP solution is a classic mathematical graph theory problem: "Given a complete graph with n vertices, each edge has a length, find the shortest closed circuit that passes through each vertex exactly once." The algorithms used to solve for the optimal path using TSP include, but are not limited to, classic algorithms such as the serpentine algorithm, greedy algorithm, Dijkstra's algorithm, genetic algorithm, and ant colony algorithm.

[0050] In step S3, after outputting the optimal path layout diagram, the layout scheme can be further refined manually to make the inspection path more reasonable.

[0051] Furthermore, in the embodiments, it can be considered that in step S1, when the average spacing between rows and columns is greater than twice the minimum recognition distance of the inspection robot, the plane coordinates of all upper chord nodes are used to form a key point coordinate set; when the average spacing between rows and columns is less than twice the minimum recognition distance of the inspection robot, a sampling method is used to extract key point coordinate values ​​from the plane coordinates of the upper chord nodes to form a key point coordinate set.

[0052] In this embodiment, the method for forming the key point coordinate set is selected based on the comparison between the average spacing of rows and columns and the minimum recognition distance of the inspection robot: one method is to directly form the key point coordinate set from all the plane coordinate sets of the upper chord nodes, and the other method is to use a sampling method to extract key point coordinate values ​​from the plane coordinate sets of the upper chord nodes to form the key point coordinate set.

[0053] Furthermore, in the embodiments, the sampling method can be considered to retain the coordinates of the outermost edge line, establish a sliding window with a length and width of twice the minimum recognition distance, and slide the row and column starting from the coordinates of the outermost edge line. Each window retains only one point, which is called a key point.

[0054] Furthermore, in the embodiments, it can also be considered that in step S1, the calculation of the net height of the spatial steel structure involves dividing the surface of the upper chord and the surface of the lower chord in the established three-dimensional solid model of the spatial steel structure into planar meshes, calculating the coordinates of the center point of each mesh, connecting the center points of the upper chord mesh and the lower chord mesh, calculating the length of each connecting line, and obtaining the net height of the spatial steel structure.

[0055] Furthermore, in the embodiments, it can be considered that in step S1, the K-means clustering algorithm is used to collect and divide the plane coordinate set of the upper chord node into X-direction coordinates and Y-direction coordinates, and calculate the average spacing of rows and columns.

[0056] Furthermore, in the embodiments, it can be considered that in step S2, the optimal path refers to the path that starts from the starting point of the inspection path, passes through each key point, and reaches the destination. It must satisfy that the key points are passed only once and the total path length is the shortest.

[0057] Furthermore, in the embodiments, it can be considered that in step S2, the TSP solution includes any one or more combinations of serpentine path algorithm, greedy algorithm, Dijkstra algorithm, genetic algorithm or ant colony algorithm.

[0058] Example 2:

[0059] An inspection robot that applies the aforementioned path planning method for tracked robot inspection of space steel structures.

[0060] Application Example 1:

[0061] Using Example 1 as an example, the application will be further explained.

[0062] like Figure 1 As shown, the method for path planning of a track robot for inspecting space steel structures according to the present invention includes the following steps:

[0063] S1: Calculate the clear height of the steel structure in each grid of the three-dimensional space to obtain the passable three-dimensional space area, convert the three-dimensional space into a two-dimensional planar grid, and calculate the set of key point coordinates.

[0064] First, based on the three data points of the inspection robot body height of 800mm, minimum safety distance of 300mm, and boom and connection height of not less than 900mm, the minimum passage height of the inspection robot is 800+300+900=2000mm=2m.

[0065] The three-dimensional solid model of the spatial steel structure was constructed using Rhino software, including the main components such as the top chord, web members, and bottom chord, as well as the plates of the top and bottom chords. The structure was divided into quadrilateral grids according to the position of the members, and the coordinates of the center point of each grid were calculated. The center points of the top chord grid and the bottom chord grid were connected, and the length of each connection was calculated, which is the net height value of different grid areas of the spatial steel structure.

[0066] Compare the net height values ​​of different grids with the minimum robot passage height of 2m obtained above and calculate the difference. Discard grids with a difference less than 0 and select the upper chord grids with a difference greater than or equal to 0, which is the range of the passable area. Calculate the planar coordinate values ​​of the upper chord nodes as a set.

[0067] Based on the analysis and testing of the selected robot's image acquisition and disease recognition capabilities, the minimum recognition distance for the inspection robot was determined to be 6m.

[0068] The set of planar coordinate values ​​of the upper chord node in S1 is approximately divided into rows and columns. That is, according to the X and Y coordinates of the coordinate values, the X and Y directions are set within a certain error. At the same time, the row and column spacing in the X and Y directions is calculated. In this embodiment, the row and column spacing is 4m, which is obtained through clustering algorithm (K-means).

[0069] The difference between the row and column spacing and twice the robot's minimum recognition distance is calculated. Since 4m is less than twice 6m, 4-12=-8, and the difference is less than 0. Therefore, a sampling method is needed to extract key points from the set of planar coordinate values ​​of the upper chord nodes. First, the lower left corner of the outermost edge is selected as the starting point, and a 12m x 12m sliding window is established. Then, the window is moved sequentially by row and column, retaining only one key point within each window. Finally, the extracted key point planar coordinate values ​​are used to form a set of point coordinates.

[0070] If the difference calculated above is greater than 0, sliding sampling can be skipped, and TSP can be solved directly.

[0071] S2: Based on the set of key point coordinates, group the solutions to obtain the optimal path and the total path length;

[0072] Taking into account factors such as the spatial steel structure's planar layout, maintenance access point setup, and inspection safety, this application example selects the lower left corner of the two-dimensional planar grid as the starting point of the inspection route.

[0073] The set of key point planar coordinate values ​​(X coordinates, Y coordinates) is transformed into a TSP problem. Using Python programming, the following algorithms are used to solve the problem: serpentine path algorithm, greedy algorithm, Dijkstra's algorithm, genetic algorithm, and ant colony algorithm. The calculation shows that the total path length obtained by the greedy algorithm is the shortest, which is 811.54m. The path has no repetition or intersection, and all key points are passed exactly once, which meets the requirements.

[0074] Software generated by Python programming, such as Figure 2 As shown, by selecting a starting point, a travel path is obtained.

[0075] Based on the selected robot's lithium battery capacity, power consumption during inspection tasks, and charging device capabilities, the maximum range of the inspection robot is determined to be 600m.

[0076] Dividing the calculated total path length of 811.54m by 600m yields k=1.35>1, therefore the key point coordinate set needs to be divided into 2 groups.

[0077] The total length was divided into two segments. Combining the coordinates of key points and considering an allowable error of no more than one row and column spacing, the total lengths of the two groups were found to be close, avoiding a large difference between the two groups. Then, the bottom left corner of the first group was selected as the starting point, and the following algorithms were used to solve the problem: serpentine path algorithm, greedy algorithm, Dijkstra's algorithm, genetic algorithm, and ant colony algorithm. The calculations showed that the greedy algorithm yielded the shortest total path length, 396.82m.

[0078] For the second group, the upper right corner was selected as the starting point, and the serpentine path algorithm, greedy algorithm, Dijkstra's algorithm, genetic algorithm, and ant colony algorithm were used to solve the problem. The calculation showed that the serpentine path algorithm yielded the shortest total path length of 402.68m, and the total path length was 799.5m, which is less than 811.54m, with a difference of 11.5m, which is no more than 12m (twice the minimum recognition distance). The two groups of paths were separated by a unit length, so the result was reasonable.

[0079] If k is not greater than 1, the grouped TSP solution can be skipped, and the optimal path and total length can be directly output.

[0080] S3: Input the coordinates of the key points corresponding to the optimal path into a three-dimensional space to verify the rationality of the planned path.

[0081] The two sets of upper chord key point coordinates, starting point coordinates, and path obtained in S2 are input into Rhino. Based on the minimum arrangement spacing of 3m, the hanging structure, connection, track, and robot are modeled in 3D solid form according to actual dimensions. The walkthrough function of BIM software such as Rhino is used to perform animation simulation to verify that the robot runs along the planned path. When no collision occurs and the minimum safety distance requirement is met, the planned path is considered reasonable.

[0082] The above embodiments have provided a detailed description of the present invention, but the content described is only a preferred embodiment of the present invention and should not be considered as limiting the scope of the present invention. All equivalent variations and improvements made within the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A track robot inspection path planning method for a spatial steel structure, characterized in that, The method comprises the following steps: S1: a three-dimensional entity model of the space steel structure is established, the net height of the space steel structure is calculated, the net height of the space steel structure is compared with the minimum passing height of the inspection robot, the region with the net height of the space steel structure not less than the minimum passing height of the inspection robot is selected as the passable region range, the outermost edge line of the passable region is confirmed, the plane coordinate values of the upper chord nodes are extracted to form a set of plane coordinate values of the upper chord nodes; The minimum recognition distance of the inspection robot is determined, the set of plane coordinate values of the upper chord nodes is approximately divided according to the rows and columns, the average spacing of the rows and columns is calculated, and a set of key point coordinate values is formed; S2: the starting point of the inspection path is preset, the optimal path is solved by TSP according to the key point coordinate values, the total path length is obtained, the total path length is divided by the longest endurance distance of the inspection robot, the integer part is taken, and the minimum number k of required inspection robots is obtained; When k>1, the set of key point coordinate values is divided into k+1 groups, the TSP of each group of key point coordinate values is solved, the optimal path of each group is obtained, and the optimal total path length of the k+1 groups is obtained by summation; When k≤1, the TSP solving is ignored, and the result of the TSP solving is directly output to obtain the optimal path and the optimal total path length; S3: the key point coordinate values of the k+1 groups, the starting point coordinate and the optimal path are input into the three-dimensional entity model, the track arrangement path is formed according to the minimum arrangement spacing of the hanging structure of the inspection robot, and the three-dimensional entity models of the hanging structure, the track and the robot are correspondingly established, and the path rationality is verified; When no collision occurs and the minimum safety distance requirement is met, it is considered that the planned path is reasonable, and the optimal path arrangement diagram is output. 2.The method for path planning of track robot inspection of spatial steel structure according to claim 1, characterized in that: In the step S1, when the average spacing of the rows and columns is greater than 2 times the minimum recognition distance of the inspection robot, the set of plane coordinate values of all the upper chord nodes is formed into the set of key point coordinate values; when the average spacing of the rows and columns is less than 2 times the minimum recognition distance of the inspection robot, the set of plane coordinate values of the upper chord nodes is extracted by a sampling method to form the set of key point coordinate values. 3.The method for path planning of track robot inspection of spatial steel structure according to claim 2, characterized in that: The sampling method is to reserve the coordinate points of the outermost edge line, establish a sliding window with a length and a width of 2 times the minimum recognition distance, start from the coordinate point of the outermost edge line, slide in the rows and columns, and only reserve one point in each window, which is called a key point. 4.The method for path planning of track robot inspection of spatial steel structure according to claim 1, characterized in that: In the step S1, the net height of the space steel structure is calculated by dividing the surfaces of the upper chord and the lower chord of the established three-dimensional entity model of the space steel structure into plane grids, calculating the coordinate of the center point of each grid, connecting the center points of the upper chord grids and the center points of the lower chord grids, and calculating the length of each connecting line to obtain the net height of the space steel structure. 5.The method for path planning of track robot inspection of spatial steel structure according to claim 1, characterized in that: In the step S1, the K-means clustering algorithm is used to collect and divide the X-direction coordinate and Y-direction coordinate of the set of plane coordinate values of the upper chord nodes, and the average spacing of the rows and columns is calculated. 6.The method for path planning of track robot inspection of spatial steel structure according to claim 1, characterized in that: In the step S2, the optimal path refers to a path from the starting point of the inspection path to the terminal point after passing through each key point, which needs to satisfy that the key point is passed only once and the total path length is the shortest. 7.The method for path planning of track robot inspection of spatial steel structure according to claim 1, characterized in that: The step S2, the TSP solution includes using any one or more combinations of a snake path algorithm, a greedy algorithm, a Dijkstra algorithm, a genetic algorithm or an ant colony algorithm.

8. An inspection robot applying the path planning method for track robot inspection of a spatial steel structure according to any one of claims 1 to 7.

Images