Unmanned aerial vehicle tunnel inspection method, device and equipment and storage medium

Abstract

The invention discloses an unmanned aerial vehicle tunnel inspection method, device and equipment and a storage medium, and belongs to the technical field of unmanned aerial vehicle inspection. The method comprises the following steps: acquiring tunnel point cloud data of a to-be-inspected tunnel; determining a tunnel skeleton structure of the to-be-inspected tunnel based on the tunnel point cloud data; based on the tunnel skeleton structure, inspection path information of the to-be-inspected tunnel is determined, so that the unmanned aerial vehicle inspects the to-be-inspected tunnel based on the inspection path information, and tunnel inspection information is generated. By using the technical scheme provided by the invention, intelligent routing inspection of the unmanned aerial vehicle in a complex tunnel environment can be realized, and the efficiency and comprehensiveness of tunnel routing inspection of the unmanned aerial vehicle are improved.

Description

Technical Field

[0001] The present invention relates to the field of drone inspection technology, and in particular to a drone tunnel inspection method, device, equipment and storage medium. Background Art

[0002] With the rapid development of modern transportation infrastructure, tunnels, as a key component, have a safety that is directly linked to the stability of public transportation and economic development. With the acceleration of urbanization and the increase in traffic volume, the frequency and importance of tunnel use have increased significantly, and traditional inspection methods are unable to meet the growing demand. Furthermore, due to the complex environment, limited space, and structural characteristics of tunnels, traditional inspection methods face numerous challenges in tunnel monitoring and maintenance. Traditional tunnel inspections rely primarily on manual labor, using visual observation and simple instrument measurements to determine the health of the tunnel structure. This method is inefficient, highly dangerous, and lacks comprehensive coverage of the tunnel area. Furthermore, manual inspections are limited by the expertise and experience of personnel, making it difficult to accurately detect potential problems such as small cracks and micro-deformations, which can easily lead to hidden dangers being overlooked.

[0003] In recent years, drones have become an increasingly important tool for tunnel inspections due to their flexibility and high degree of automation. Drones equipped with sensors (such as lidar and high-definition cameras) can collect high-precision tunnel data, overcoming the limitations of manual inspections. However, existing technologies still have shortcomings in many areas. Specifically, most drone inspection path planning methods use simple preset models and fail to fully consider the complexities of tunnel structures, resulting in inaccurate path planning and low inspection efficiency. Furthermore, due to inappropriate drone flight path design, blind spots may exist, preventing comprehensive tunnel coverage.

[0004] Therefore, a more reliable solution needs to be provided. Summary of the Invention

[0005] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method, device, equipment and storage medium for drone tunnel inspection, which can realize intelligent inspection of drones in complex tunnel environments and improve the efficiency and comprehensiveness of drone tunnel inspections.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] In one aspect, the present invention provides a method for tunnel inspection using a drone, the method comprising:

[0008] Obtain tunnel point cloud data of the tunnel to be inspected;

[0009] Determining the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data;

[0010] Based on the tunnel skeleton structure, the inspection path information of the tunnel to be inspected is determined, so that the drone inspects the tunnel to be inspected based on the inspection path information and generates tunnel inspection information.

[0011] In some possible implementations, determining the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data includes:

[0012] Determining the point cloud density of each point based on the tunnel point cloud data;

[0013] According to the point cloud density of each point, each point is screened to obtain a candidate skeleton point; the candidate skeleton point is a point whose point cloud density is greater than or equal to a preset density threshold;

[0014] Based on the candidate skeleton points and in combination with a preset skeleton extraction algorithm, the tunnel skeleton structure is determined.

[0015] In some possible implementations, determining the inspection path information of the tunnel to be inspected based on the tunnel skeleton structure includes:

[0016] vertically slicing the tunnel skeleton structure along the skeleton line to determine a plurality of skeleton cutting points;

[0017] Determining multiple inspection points of the tunnel to be inspected according to the multiple skeleton tangent points;

[0018] The inspection path information is determined based on the multiple inspection points.

[0019] In some possible implementations, determining the multiple inspection points of the tunnel to be inspected based on the multiple skeleton tangent points includes:

[0020] Taking each skeleton cut point as the center and combining it with the sampling radius, multiple sampling areas are determined;

[0021] According to each sampling area, determining a set of sampling points corresponding to each sampling area in the tunnel point cloud data;

[0022] The sampling points in each sampling point set are moved along a target direction by a preset distance to obtain a plurality of moved sampling points, and the plurality of moved sampling points are used as the plurality of inspection points.

[0023] In some possible implementations, determining the inspection path information based on the multiple inspection points includes:

[0024] Based on the multiple inspection points, construct an inspection path determination problem;

[0025] Based on a preset path determination algorithm, the inspection path determination problem is solved to determine the inspection path information.

[0026] In some possible implementations, constructing an inspection path determination problem based on the multiple inspection points includes:

[0027] Determining the plurality of inspection points as a plurality of vertices of a path graph to be determined;

[0028] Connecting each vertex to obtain multiple edges of the path graph to be determined;

[0029] Determine the edge weight corresponding to each edge based on the distance between each vertex;

[0030] Constructing the to-be-determined path graph according to a plurality of vertices of the to-be-determined path graph, a plurality of edges of the to-be-determined path graph, and an edge weight corresponding to each edge;

[0031] Based on the path graph to be determined, the inspection path determination problem is constructed; the inspection path determination problem is to determine the shortest path traversing to each vertex of the path graph to be determined; wherein, the constraint conditions of the inspection path determination problem are that each vertex of the path graph to be determined is visited only once; and the path that can traverse to each vertex of the path graph to be determined has no self-intersection.

[0032] In some possible implementations, solving the inspection path determination problem based on a preset path determination algorithm and determining the inspection path information includes:

[0033] Setting a trace strength for each edge of the path graph to be determined; wherein the trace strength is reduced proportionally;

[0034] Set a preset number of path exploration bodies;

[0035] Randomly selecting a vertex from among the multiple vertices of the to-be-determined path graph as a starting point through each path exploration body;

[0036] and determining a current path exploration point among a plurality of vertices of the to-be-determined path graph based on the trace strength;

[0037] and determining a current exploration path based on the current path exploration point;

[0038] and in the process of the path exploration body exploring the current exploration path, updating the trace strength corresponding to the current exploration path based on the edge weight of the current exploration path, jumping to the step of determining the current path exploration point among the multiple vertices of the path graph to be determined based on the trace strength, until the path exploration body completes the exploration of the multiple vertices of the path graph to be determined; determining the exploration completion path of the path exploration body, and updating the path trace strength of the exploration completion path;

[0039] Returning to each path exploration volume, randomly selecting a vertex from the multiple vertices of the path graph to be determined as a starting point until the number of iterations reaches a preset number;

[0040] The exploration completed path whose path trace intensity is greater than a preset intensity threshold is determined as the inspection path information.

[0041] Another aspect provides a UAV tunnel inspection device, the device comprising:

[0042] A point cloud data acquisition module is used to acquire tunnel point cloud data of the tunnel to be inspected;

[0043] A skeleton structure determination module, configured to determine the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data;

[0044] The inspection path determination module is used to determine the inspection path information of the tunnel to be inspected based on the tunnel skeleton structure, so that the drone can inspect the tunnel to be inspected based on the inspection path information and generate tunnel inspection information.

[0045] On the other hand, an electronic device is provided, which includes a processor and a memory, wherein the memory stores at least one instruction and at least one program, and the at least one instruction and the at least one program are loaded and executed by the processor to implement the drone tunnel inspection method as described above.

[0046] On the other hand, a computer-readable storage medium is provided, in which at least one instruction and at least one program are stored. The at least one instruction and the at least one program are loaded and executed by a processor to implement the drone tunnel inspection method as described above.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] In the present invention, by obtaining tunnel point cloud data of the tunnel to be inspected; based on the tunnel point cloud data, the tunnel skeleton structure of the tunnel to be inspected is determined, and the corresponding tunnel skeleton structure can be obtained for tunnels of different geometric shapes, thereby improving the accuracy of determining the tunnel skeleton structure; then, based on the tunnel skeleton structure, the inspection path information of the tunnel to be inspected is determined, which can improve the accuracy and efficiency of determining the inspection path, obtain the inspection path information that enables the UAV to inspect the tunnel efficiently, and make the inspection coverage comprehensive and without blind spots; then, the UAV inspects the tunnel to be inspected based on the inspection path information and generates tunnel inspection information, which can improve the efficiency and accuracy of the UAV tunnel inspection, and can realize the autonomous comprehensive inspection and intelligent inspection of tunnels with different structures or different environmental types by the UAV. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 This is a flow chart of a method for tunnel inspection using a drone provided by an embodiment of the present invention;

[0051] Figure 2 This is a schematic diagram of a process for determining a tunnel skeleton structure provided by an embodiment of the present invention;

[0052] Figure 3 is a schematic diagram of a tunnel skeleton structure provided by an embodiment of the present invention;

[0053] Figure 4 is a schematic diagram of another tunnel skeleton structure provided by an embodiment of the present invention;

[0054] Figure 5 is a schematic diagram of a determined inspection point provided by an embodiment of the present invention;

[0055] Figure 6 This is a flow chart of solving the inspection path determination problem and determining inspection path information based on a preset path determination algorithm provided by an embodiment of the present invention;

[0056] Figure 7 This is a schematic diagram of inspection path information provided by an embodiment of the present invention;

[0057] Figure 8 1 is a front view of a drone provided by an embodiment of the present invention;

[0058] Figure 9is a side view of a drone provided by an embodiment of the present invention;

[0059] Figure 10 It is a structural schematic diagram of a UAV tunnel inspection device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0061] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0062] In the embodiments of the present invention, the term "module" or "unit" refers to a computer program or portion of a computer program that has a predetermined function and works together with other related components to achieve a predetermined goal. The term "module" or "unit" refers to a computer program or portion of a computer program that has a predetermined function and works together with other related components to achieve a predetermined goal. The term "module" or "unit" may be implemented in whole or in part using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a single processor (or multiple processors or memories) may be used to implement one or more modules or units. Furthermore, each module or unit may be part of an overall module or unit that incorporates the functionality of that module or unit.

[0063] Various exemplary embodiments, features, and aspects of the present invention will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

[0064] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0065] The term "and / or" herein simply describes an association relationship between associated objects, indicating that three relationships can exist. For example, "A and / or B" can represent the existence of three situations: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" herein refers to any combination of at least two of any one or more of a plurality of items. For example, "at least one of A, B, and C" can represent any one or more elements selected from the set consisting of A, B, and C.

[0066] In addition, numerous specific details are provided in the following detailed description to better illustrate the present invention. Those skilled in the art will appreciate that the present invention may be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of the present invention.

[0067] Figure 1 It is a flowchart of a method for tunnel inspection by a drone provided by an embodiment of the present invention. This specification provides method operation steps such as the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many, and does not represent the only execution order. When the actual system or server product is executed, it can be executed in sequence or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the method shown in the embodiment or the accompanying drawings. Specifically, Figure 1 As shown, the above method may include:

[0068] S101: Acquire tunnel point cloud data of the tunnel to be inspected;

[0069] In a specific embodiment, the tunnel to be inspected can include any tunnel requiring inspection. Tunnel point cloud data can represent the three-dimensional spatial data of the tunnel to be inspected. Tunnel point cloud data can record the surface information of the tunnel to be inspected in the form of points. Optionally, the tunnel point cloud data can include spatial position information of multiple points, namely, the three-dimensional coordinates of the multiple points, namely, the x-coordinate, y-coordinate, and z-coordinate. Tunnel point cloud data is highly accurate and dense, facilitating the determination of the tunnel skeleton structure of the tunnel to be inspected, thereby improving the accuracy of the tunnel skeleton structure determination. Optionally, the tunnel point cloud data can be collected using a lidar onboard an unmanned aerial vehicle system.

[0070] In a specific embodiment, before obtaining tunnel point cloud data for a tunnel to be inspected, the method may include: obtaining raw tunnel point cloud data for the tunnel to be inspected; and preprocessing the raw tunnel point cloud data to generate the tunnel point cloud data. Optionally, preprocessing the raw tunnel point cloud data may include performing operations such as denoising and filtering on the raw tunnel point cloud data to eliminate errors caused by acquisition equipment accuracy or the acquisition environment, thereby improving the quality of the point cloud data.

[0071] S102: Determine the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data;

[0072] In a specific embodiment, the tunnel skeleton structure can represent the tunnel spatial topology information of the tunnel to be inspected, and the tunnel spatial topology information may include the tunnel geometry, tunnel spatial position, etc.; optionally, the tunnel skeleton structure can be used to provide a basis for inspection path planning for drone inspection of the tunnel.

[0073] In an optional embodiment, Figure 2 : is a schematic diagram of a process for determining a tunnel skeleton structure provided by an embodiment of the present invention; Figure 2 As shown, the above-mentioned determination of the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data may include:

[0074] S201: Determine the point cloud density of each point based on the tunnel point cloud data;

[0075] S202: Screening each point according to its point cloud density to obtain a candidate skeleton point; a candidate skeleton point is a point whose point cloud density is greater than or equal to a preset density threshold;

[0076] S203: Determine the tunnel skeleton structure based on the candidate skeleton points and in combination with a preset skeleton extraction algorithm.

[0077] In a specific embodiment, the point cloud density may represent the density of points within a preset area of the point; optionally, the point cloud density of each point may be determined using the following formula:

[0078] ;

[0079] in, represents the point cloud density of point i; Represents the spatial position information of point i in the tunnel point cloud data; Represents the spatial position information of point j in the tunnel point cloud data; represents the Euclidean distance between point i and point j; h represents the bandwidth parameter of the density kernel function; K represents the Gaussian kernel function.

[0080] Optionally, the point cloud density of each point can also be determined using the following formula:

[0081] ;

[0082] in, represents the point cloud density of point i, N(i) represents the neighborhood point set of point i, and r represents the neighborhood radius. Specifically, the point cloud density can be determined based on the number of points in a preset area, and the neighborhood radius can be determined based on the actual application. Specifically, r = 0.5m.

[0083] Optionally, in the process of determining the point cloud density of each point, a K-dimensional tree or an octree (data structure) can be used to spatially partition the tunnel point cloud data, and its efficient neighborhood query capability can be used to quickly count the number of neighboring points of each point to accelerate the point cloud density search process and improve the efficiency of point cloud density determination.

[0084] In a specific embodiment, the candidate skeleton points may be key points for determining the tunnel skeleton; based on a preset density threshold, points with a point cloud density less than the preset density threshold may be removed, and points with a point cloud density greater than or equal to the preset density threshold may be screened out; optionally, the preset density threshold may be set based on actual applications. Specifically, the preset density threshold may be 10, and the point cloud density greater than or equal to the preset density threshold (i.e., ) candidate skeleton points.

[0085] In a specific embodiment, the preset path generation algorithm can be set in combination with actual applications. Optionally, the preset skeleton extraction algorithm can be a minimum spanning tree algorithm. Based on the candidate skeleton points, the minimum spanning tree algorithm is used to determine the optimal connection path between the candidate skeleton points and determine the tunnel skeleton structure. Optionally, the above-mentioned determination of the tunnel skeleton structure based on the candidate skeleton points in combination with the preset skeleton extraction algorithm may include: extracting a preliminary tunnel skeleton structure based on the candidate skeleton points in combination with the preset skeleton extraction algorithm; performing a refinement operation on the preliminary tunnel skeleton structure and removing redundant structures to obtain a modified preliminary tunnel skeleton structure; applying morphological operations to the modified preliminary tunnel skeleton structure to determine the tunnel skeleton structure; optionally, the applied morphological operations may include pruning, disconnection, etc. The above-mentioned extraction of the tunnel skeleton can ensure the correctness and rationality of the tunnel skeleton topological structure, and can optimize the integrity and accuracy of the tunnel skeleton, so as to accurately reflect the true connection relationship and global morphology of the tunnel. Optionally, Figure 3 is a schematic diagram of a tunnel skeleton structure provided by an embodiment of the present invention; Figure 3 As shown, the small dots in the figure can be candidate skeleton points, and the solid dots in the figure and the connections between the solid dots constitute the tunnel skeleton structure.

[0086] In a specific embodiment, the minimum spanning tree is given two points in the tunnel point cloud data. and The edge weight , express and The Euclidean distance between the candidate skeleton points is used to determine the optimal connection path between the candidate skeleton points using the minimum spanning tree algorithm. The corresponding objective function can be expressed as follows:

[0087]

[0088] Among them, MST represents the spanning tree with the minimum sum of weights; express and The weight between Represents the edge set of the spanning tree; specifically, the MST can be constructed using Prim's algorithm or Kruskal's algorithm to ensure that all nodes are connected with the minimum sum of weights and extract the continuous skeleton structure of the tunnel.

[0089] S103: Based on the tunnel skeleton structure, determine the inspection path information of the tunnel to be inspected, so that the drone inspects the tunnel to be inspected based on the inspection path information and generates tunnel inspection information.

[0090] In a specific embodiment, the inspection path information of the tunnel to be inspected may be a path used to inspect the tunnel to be inspected; the tunnel inspection information may be information collected by a drone for evaluating the tunnel structure, tunnel environmental conditions, etc.

[0091] In an optional embodiment, the above-mentioned determination of the inspection path information of the tunnel to be inspected based on the tunnel skeleton structure may include:

[0092] Perform vertical slicing of the tunnel skeleton structure along the skeleton line and determine multiple skeleton cutting points;

[0093] Determine multiple inspection points of the tunnel to be inspected based on multiple skeleton tangent points;

[0094] Based on multiple inspection points, inspection path information is determined.

[0095] In a specific embodiment, the skeleton line can be a continuous spatial curve running through the entire length of the tunnel and a central curve running through the length direction of the tunnel. The skeleton line can be approximately equal to the distance between each boundary of the tunnel inner wall. Specifically, Figure 4 is a schematic diagram of the skeleton line provided by an embodiment of the present invention, such as Figure 4 As shown, the candidate skeleton points for determining the tunnel skeleton structure on the skeleton line are approximately evenly distributed. The skeleton tangent point can be the intersection of the slicing plane obtained by vertically slicing the tunnel skeleton structure along the skeleton line and the skeleton line. Specifically, the skeleton tangent point can be the center point of the slicing plane.

[0096] In one specific embodiment, the tunnel skeleton structure is vertically sliced along the skeleton line to generate a series of evenly spaced slices that cover the entire area of the tunnel to be inspected. This facilitates inspections based on the determined inspection points to cover the entire tunnel area, thereby improving the comprehensiveness of tunnel inspections. Specifically, the slice interval can be set to 2m, and the tunnel skeleton structure can be vertically sliced along the skeleton line to generate 250 slices. Optionally, if the tunnel to be inspected has a complex shape, the distance between slices can be appropriately reduced at corners or irregular areas to improve the effectiveness of inspection point determination and avoid inspection blind spots.

[0097] In an optional embodiment, the above-mentioned determination of multiple inspection points of the tunnel to be inspected based on multiple skeleton tangent points may include:

[0098] Taking each skeleton cut point as the center and combining it with the sampling radius, multiple sampling areas are determined;

[0099] According to each sampling area, a set of sampling points corresponding to each sampling area is determined in the tunnel point cloud data;

[0100] The sampling points in each sampling point set are moved along the target direction by a preset distance to obtain a plurality of moved sampling points, and the plurality of moved sampling points are used as a plurality of inspection points.

[0101] In a specific embodiment, the sampling radius can be set in combination with actual applications. Optionally, the sampling radius can cover all points of the tunnel section to be inspected. Specifically, the sampling radius can be determined by the following formula: Where R represents the sampling radius, D represents the cross-sectional diameter of the tunnel to be inspected, and m represents the radius redundancy distance. Optionally, the radius redundancy distance can be set based on actual application needs. Specifically, the radius redundancy distance can be 0.1m-0.2m. Optionally, setting the radius redundancy distance can adapt to different tunnel structures and improve the accuracy of sampling radius determination, thereby improving the effectiveness of inspection point determination.

[0102] The sampling area can be an area used to obtain sampling points and subsequently determine inspection points; optionally, the sampling area can be a circular area generated with the skeleton tangent point as its center. The sampling point set can include multiple sampling points. The sampling points can be points within the sampling area and can be used to determine inspection points. Furthermore, the sampling points can be points selected from the tunnel point cloud data whose distance from the skeleton tangent point is less than the sampling radius. Optionally, a two-dimensional projection can be performed on the sampling point set of each determined sampling area to form a planar slice image, which can be used for tunnel cross-section analysis or inspection point determination. Inspection points can be points used to determine the inspection path of the tunnel to be inspected.

[0103] In a specific embodiment, the target direction can be set in combination with the actual application, and optionally, the target direction can be the direction toward the skeleton tangent point. The preset distance can be set in combination with the actual application, and optionally, the preset distance can be determined according to the following formula: ; Where d represents the preset distance, D represents the cross-sectional diameter of the tunnel to be inspected, Indicates the horizontal viewing angle of the camera carried by the drone. When the vertical viewing angle of the camera reaches In this case, it is necessary to adjust the camera angle or increase the preset distance to ensure the accuracy of the inspection point determination.

[0104] In a specific embodiment, Figure 5 Schematic diagram of the inspection points determined according to the embodiment of the present invention; Figure 5 As shown, 15 is the skeleton tangent point, 16 is the sampling point, and 17 is the inspection point. The inspection point can be obtained by moving the sampling point toward the skeleton tangent point by a preset distance, that is, the inspection point can be obtained by retreating the sampling point by a preset distance. Specifically, the figure only shows the inspection points determined in one of the tunnel cross-sectional slice planes, and for the convenience of explanation, not all inspection points are shown. The specific inspection points, sampling points and skeleton tangent points are determined in combination with actual applications.

[0105] In an optional embodiment, the above-mentioned determination of inspection path information based on multiple inspection points may include:

[0106] Based on multiple inspection points, construct the inspection path determination problem;

[0107] Based on the preset path determination algorithm, the inspection path determination problem is solved and the inspection path information is determined.

[0108] In a specific embodiment, the inspection path determination problem can be used to determine the optimal path for inspecting a tunnel to be inspected.

[0109] In an optional embodiment, the above-mentioned problem of constructing an inspection path determination problem based on multiple inspection points may include:

[0110] Determine multiple inspection points as multiple vertices of the path graph to be determined;

[0111] Connect each vertex to obtain multiple edges of the path graph to be determined;

[0112] Based on the distance between each vertex, determine the edge weight corresponding to each edge;

[0113] Constructing a path graph to be determined according to a plurality of vertices of the path graph to be determined, a plurality of edges of the path graph to be determined, and an edge weight corresponding to each edge;

[0114] Based on the path graph to be determined, a patrol path determination problem is constructed; the patrol path determination problem is to determine the shortest path traversing to each vertex of the path graph to be determined; wherein, the constraints of the patrol path determination problem are that each vertex of the path graph to be determined is visited only once; and the path that can traverse to each vertex of the path graph to be determined has no self-intersection.

[0115] In a specific embodiment, the to-be-determined path graph may be used to plan the shortest path for inspecting the tunnel to be inspected.

[0116] In an optional embodiment, Figure 6 This is a flow chart of solving the inspection path determination problem and determining the inspection path information based on a preset path determination algorithm provided by an embodiment of the present invention; Figure 6 As shown, the above-mentioned solution to the inspection path determination problem based on the preset path determination algorithm and determination of the inspection path information may include:

[0117] S601: Setting a trace strength for each edge of the path graph to be determined; wherein the trace strength is reduced proportionally;

[0118] S602: Setting a preset number of path exploration bodies;

[0119] S603: randomly selecting a vertex from multiple vertices of the path graph to be determined as a starting point through each path exploration body;

[0120] S6031: and determining a current path exploration point among a plurality of vertices of the path graph to be determined based on the trace strength;

[0121] S6032: Determine the current exploration path based on the current path exploration point;

[0122] S6033: During the process of the path exploration body exploring the current exploration path, based on the edge weights of the current exploration path, the trace strength corresponding to the current exploration path is updated, and the process proceeds to step S6031 until the path exploration body completes the exploration of multiple vertices of the path graph to be determined; the exploration completion path of the path exploration body is determined, and the path trace strength of the exploration completion path is updated;

[0123] S6034: Return to S603 until the number of iterations reaches the preset number;

[0124] S604: Determine the explored path whose path trace strength is greater than a preset strength threshold as inspection path information.

[0125] In a specific embodiment, trace strength can represent the exploration effectiveness of an exploration path (edges between vertices in the path graph to be determined). A high trace strength indicates that the exploration path is selected by more path explorers and is closer to solving the inspection path determination problem; a low trace strength indicates that the exploration path performs poorly. Initially setting a trace strength for each edge in the path graph to be determined can encourage path explorers to explore paths.

[0126] The path exploration body can be used to traverse each vertex in the path graph to be determined to determine the exploration path and then determine the inspection path information. The current path exploration point can be the end point of the path to be explored by the path exploration body; the current exploration path can be the path that the path exploration body currently wants to explore. Specifically, the current exploration path can be the path between the starting point and the current path exploration point. In the process of the path exploration body exploring the current exploration path, the trace strength corresponding to the current exploration path is generated based on the edge weight of the current exploration path. At this time, the end point of the current exploration path, that is, the current path exploration point, becomes a historical path exploration point. The historical path exploration point can be used as the starting point. Then, based on the trace strength, the current path exploration point is determined among the multiple vertices in the path graph to be determined.

[0127] The exploration completion path can be the path after the path exploration body completes the exploration of each vertex in the path graph to be determined according to the edge weights and trace strength of the path graph to be determined; the path trace strength can represent the trace strength of the exploration completion path after the path exploration body completes the path exploration. The number of iterations reaching a preset number can be a preset iteration convergence condition, and the preset number can be set in combination with actual applications. Optionally, when determining the exploration completion path of the path exploration body and updating the path trace strength of the exploration completion path, the length of the exploration completion path is updated and recorded; after each iteration, the lengths of all recorded exploration completion paths can be compared to determine and update the shortest exploration completion path; at this time, the preset iteration convergence condition can be that the shortest exploration completion path remains unchanged for multiple consecutive times.

[0128] The preset strength threshold can be set based on actual application. Explored paths with path trace strength greater than the preset strength threshold are determined as inspection path information. Specifically, the explored path with the highest path trace strength is determined as the optimal path for inspecting the tunnel to be inspected. Optionally, the path trace strength of the explored paths is proportionally reduced.

[0129] In a specific embodiment, the preset path determination algorithm can be set in combination with actual applications. Specifically, the preset path determination algorithm can be an ant colony algorithm, and accordingly, the trace strength can be a pheromone concentration. By utilizing the ant colony algorithm, the patrol path determination problem is solved, and the pheromone concentration on the path is dynamically updated, so that the ants tend to the shortest path to determine the patrol path information. Optionally, in the process of solving the patrol path determination problem based on the preset path determination algorithm, it is necessary to combine the constraints of the patrol path determination problem. The constraint is that each vertex of the path graph to be determined is visited only once; the path that can traverse to each vertex of the path graph to be determined has no self-intersection. Optionally, the trace strength of each explored path and the trace strength of the explored completed path are reduced proportionally to simulate the volatilization of pheromones over time in nature, and the pheromones on the path will be reduced proportionally.

[0130] In a specific embodiment, the current path exploration point is determined among multiple vertices of the path graph to be determined based on the trace strength. Alternatively, the current path exploration point is determined among multiple vertices of the path graph to be determined based on the trace strength and the edge weights of the path graph to be determined. Specifically, the selection probability of the path exploration point can be determined based on the following formula:

[0131] P ij = [ T ij ] α [ η ij ] β ∑ k ∈ Allow Node [ T ik ] α [ η ik ] β ;

[0132] in, represents the probability of selecting from node i to node j in the path graph to be determined; represents the trail strength (pheromone concentration) of the explored path; represents the heuristic information of the exploration path, represents the distance from node i to node j; represents the weight parameter that controls the trace strength; Indicates the weight parameter for controlling heuristic information. Optionally, the weight parameter for controlling trace intensity and the weight parameter for controlling heuristic information can be set in combination with actual applications. Specifically, , .

[0133] In a specific embodiment, Figure 7 is a schematic diagram of inspection path information provided by an embodiment of the present invention; Figure 7 As shown in Figure 1, the optimal inspection path planned for the tunnel to be inspected is shown, that is, the shortest path that can traverse to each inspection point. Figure 7 This is a schematic diagram to illustrate the inspection path information. Figure 5 The inspection points shown are not directly related.

[0134] In the above embodiment, based on the preset path algorithm, the inspection path determination problem is solved and the inspection path information is determined, which can improve the accuracy and effectiveness of the inspection path, and then improve the efficiency and comprehensiveness of tunnel inspection, thereby ensuring the accuracy, comprehensiveness and effectiveness of tunnel inspection information.

[0135] In a specific embodiment, after the inspection path information is determined, the drone can inspect the tunnel to be inspected according to the inspection path information and generate tunnel inspection information, thereby completing tunnel inspection based on the drone, which can improve the accuracy and efficiency of tunnel inspection and improve the safety of staff.

[0136] An embodiment of the present invention further provides a drone, which may include a sensing module, a computing module, a flight control module, a power supply module, and a shock absorption module;

[0137] A perception module is used to collect tunnel information during the inspection of the tunnel to be inspected. The tunnel information may include tunnel point cloud data and tunnel images. Specifically, the perception module may include a laser radar and a camera. The laser radar may be used to collect tunnel point cloud data, and the camera may be used to obtain high-resolution tunnel images and support SLAM algorithms.

[0138] Computing module: The embedded computing unit integrates SLAM algorithm, preset skeleton extraction algorithm, preset path determination algorithm and obstacle avoidance logic for real-time data processing and navigation control;

[0139] The flight control module is used to collaborate with the computing module to control the drone's precise flight in various environments. Specifically, the flight control module can include a flight controller. A customized controller can be used within the flight control module to provide more stable environmental modeling capabilities and adjustable motion optimization. The software can be built on the Robot Operating System (ROS) Noetic and run on Ubuntu 20.04. The Pixhack-6c flight controller communicates with the onboard computer via the digital communication protocol Dshot600, the electronic speed controller (ESC), and a universal asynchronous receiver / transmitter (UART).

[0140] The power supply module is used to provide power to the drone. Specifically, the drone adopts a lightweight design and can be equipped with four SUNNYSKY X-2216 II KV1280 brushless DC motors controlled by the XF-35Aesc. The power supply module adopts a modular quick-release design to support long-term flight and rapid replacement.

[0141] The shock-absorbing module is used to improve the accuracy of data collection. Specifically, the shock-absorbing module may include a shock-absorbing bracket and a special shock-absorbing layer. The shock-absorbing bracket is used to be installed on the laser radar and camera to reduce the vibration of the laser radar and camera. The special shock-absorbing layer is used to stabilize the flight control module and the computing module, thereby improving the measurement accuracy during the flight.

[0142] Specifically, Figure 8 1 is a front view of a drone provided by an embodiment of the present invention. Figure 9 : is a side view of a drone provided by an embodiment of the present invention; Figure 8 and Figure 9 As shown, the UAV includes a laser radar 1, which can be used to collect tunnel point cloud data of the tunnel to be inspected. Specifically, the laser radar can be a LIVOX-MID360 laser radar, which can detect the environment 360° without blind spots and collect point cloud data; a radar fixing bracket 2, which is used to place, support and fix the laser radar 1; a camera 3, which is used to collect images of the tunnel to be inspected during the UAV inspection of the tunnel to be inspected. Specifically, the camera can be a RealSense D435 depth camera; computing unit 4, used to process collected point cloud data and images; wing plate 5, used to support the flight assembly, which may include a motor 6 for powering the rotor 7, and the rotor 7 for generating lift and thrust; an electric speed controller 8, used to adjust the power output of the motor 6; an electric speed controller board 9, used to fix the electric speed controller and provide a shock absorption function; a flight controller 10, used to control the attitude of the UAV and determine the positioning and navigation of the UAV according to the inspection path information; a flight control fixing layer 11, used to stabilize the flight controller 10 and reduce vibration; a battery 12, used to provide power to the UAV; a battery board 13, used to fix the battery 12 and provide bottom balance for the battery 12; a gimbal 14, used to carry the camera 3, which can dynamically adjust the attitude to adjust the camera viewing angle, thereby ensuring the stability and accuracy of data acquisition.

[0143] In a specific embodiment, combined with the inspection path information, the pan-tilt control system can adjust the posture of the pan-tilt in real time through a control algorithm to adjust the angle of the camera, thereby ensuring that the camera can always be aimed at the target area, that is, the tunnel wall corresponding to the inspection path information; the computing unit processes the tunnel point cloud data collected by the lidar and the tunnel image collected by the camera in real time to determine the actual situation in the tunnel, and then determine whether the inspection path information needs to be adjusted, and if the inspection path information needs to be adjusted, dynamic adjustment information of the inspection path information can be sent to the pan-tilt control system to deal with emergencies inside the tunnel, such as obstacles or changes in ambient light.

[0144] In a specific embodiment, the gimbal control system takes into account the changes in the position, speed, and distance between inspection points of the drone when adjusting the gimbal posture, thereby ensuring the continuity and accuracy of the image and point cloud data. The specific adjustment can be based on the following formula:

[0145] , ;

[0146] in, Indicates the horizontal angle of the gimbal. Indicates the vertical angle of the gimbal; , and Indicates the real-time location coordinates of the inspection point, Indicates the distance between the drone and the target point, which can be the next inspection point corresponding to the current inspection point.

[0147] It can be seen from the technical solutions provided in the above embodiments of this specification that this specification obtains tunnel point cloud data of the tunnel to be inspected; based on the tunnel point cloud data, the tunnel skeleton structure of the tunnel to be inspected is determined, and the corresponding tunnel skeleton structure can be obtained for tunnels of different geometric shapes, thereby improving the accuracy of determining the tunnel skeleton structure; then, based on the tunnel skeleton structure, the inspection path information of the tunnel to be inspected is determined, which can improve the accuracy and efficiency of determining the inspection path, and obtain inspection path information that enables UAVs to inspect tunnels efficiently, and can make the inspection coverage comprehensive and without blind spots; then, based on the inspection path information, the UAV inspects the tunnel to be inspected and generates tunnel inspection information, which can improve the efficiency and accuracy of UAV tunnel inspection, and can realize autonomous comprehensive inspection and intelligent inspection of tunnels with different structures or different environmental types by UAVs.

[0148] The embodiment of the present invention also provides a UAV tunnel inspection device, correspondingly, Figure 10 : is a structural diagram of a UAV tunnel inspection device provided by an embodiment of the present invention; Figure 10 As shown, the above device includes:

[0149] The point cloud data acquisition module 1010 is used to acquire tunnel point cloud data of the tunnel to be inspected;

[0150] A skeleton structure determination module 1020 is configured to determine the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data;

[0151] The inspection path determination module 1030 is used to determine the inspection path information of the tunnel to be inspected based on the tunnel skeleton structure, so that the drone can inspect the tunnel to be inspected based on the inspection path information and generate tunnel inspection information.

[0152] In an optional embodiment, the skeleton structure determination module 1020 includes:

[0153] a point cloud density determining unit, configured to determine the point cloud density of each point based on the tunnel point cloud data;

[0154] a candidate skeleton point determination unit, configured to screen each point according to the point cloud density of each point to obtain a candidate skeleton point; the candidate skeleton point is a point whose point cloud density is greater than or equal to a preset density threshold;

[0155] The skeleton structure determining unit is configured to determine the tunnel skeleton structure based on the candidate skeleton points in combination with a preset skeleton extraction algorithm.

[0156] In an optional embodiment, the inspection path determination module 1030 includes:

[0157] a skeleton tangent point determination module, configured to perform vertical slicing processing on the tunnel skeleton structure along the skeleton line to determine a plurality of skeleton tangent points;

[0158] An inspection point determination module, configured to determine a plurality of inspection points of the tunnel to be inspected based on the plurality of skeleton tangent points;

[0159] The path information determination module is configured to determine the inspection path information based on the multiple inspection points.

[0160] In an optional embodiment, the inspection point determination module includes:

[0161] A sampling region determination unit is used to determine multiple sampling regions based on each skeleton cut point and a sampling radius;

[0162] a sampling point determination unit, configured to determine, based on each sampling area, a set of sampling points corresponding to each sampling area in the tunnel point cloud data;

[0163] The inspection point determination unit is configured to move the sampling points in each sampling point set along a target direction by a preset distance to obtain a plurality of moved sampling points, and use the plurality of moved sampling points as the plurality of inspection points.

[0164] In an optional embodiment, the path information determination module includes:

[0165] a path determination problem construction unit, configured to construct an inspection path determination problem based on the plurality of inspection points;

[0166] The path information determination unit is used to solve the inspection path determination problem based on a preset path determination algorithm and determine the inspection path information.

[0167] In an optional embodiment, the path determination problem construction unit is specifically used to

[0168] Determining the plurality of inspection points as a plurality of vertices of a path graph to be determined;

[0169] Connecting each vertex to obtain multiple edges of the path graph to be determined;

[0170] Determine the edge weight corresponding to each edge based on the distance between each vertex;

[0171] Constructing the to-be-determined path graph according to a plurality of vertices of the to-be-determined path graph, a plurality of edges of the to-be-determined path graph, and an edge weight corresponding to each edge;

[0172] Based on the path graph to be determined, the inspection path determination problem is constructed; the inspection path determination problem is to determine the shortest path traversing to each vertex of the path graph to be determined; wherein, the constraint conditions of the inspection path determination problem are that each vertex of the path graph to be determined is visited only once; and the path that can traverse to each vertex of the path graph to be determined has no self-intersection.

[0173] In an optional embodiment, the path information determination unit is specifically configured to:

[0174] Setting a trace strength for each edge of the path graph to be determined; wherein the trace strength is reduced proportionally;

[0175] Set a preset number of path exploration bodies;

[0176] Randomly selecting a vertex from among the multiple vertices of the to-be-determined path graph as a starting point through each path exploration body;

[0177] and determining a current path exploration point among a plurality of vertices of the to-be-determined path graph based on the trace strength;

[0178] and determining a current exploration path based on the current path exploration point;

[0179] and in the process of the path exploration body exploring the current exploration path, updating the trace strength corresponding to the current exploration path based on the edge weight of the current exploration path, jumping to the step of determining the current path exploration point among the multiple vertices of the path graph to be determined based on the trace strength, until the path exploration body completes the exploration of the multiple vertices of the path graph to be determined; determining the exploration completion path of the path exploration body, and updating the path trace strength of the exploration completion path;

[0180] Returning to each path exploration volume, randomly selecting a vertex from the multiple vertices of the path graph to be determined as a starting point until the number of iterations reaches a preset number;

[0181] The exploration completed path whose path trace intensity is greater than a preset intensity threshold is determined as the inspection path information.

[0182] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0183] In a specific embodiment, a UAV tunnel inspection device can be set on a UAV. The UAV tunnel inspection device can obtain the tunnel point cloud data of the tunnel to be inspected collected by the UAV laser radar, and then determine the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data. Then, based on the tunnel skeleton structure, determine the inspection path information of the tunnel to be inspected, and then send the inspection path information to the flight control module of the UAV. The flight control module of the UAV controls the flight of the UAV based on the inspection path information to inspect the tunnel to be inspected.

[0184] An embodiment of the present invention also provides an electronic device, comprising: a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the drone tunnel inspection method as described in any one of the method embodiments.

[0185] An embodiment of the present invention also provides a computer storage medium, which can be set in a server to store at least one instruction, at least one program, code set or instruction set for implementing the method embodiment. The at least one instruction, the at least one program, the code set or instruction set is loaded and executed by the processor to implement the drone tunnel inspection method as described in any one of the method embodiments.

[0186] Optionally, in an embodiment of the present invention, the storage medium may be located in at least one of a plurality of network servers in a computer network. Optionally, in an embodiment of the present invention, the storage medium may include, but is not limited to, a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard drive, a magnetic disk, or an optical disk, among other media capable of storing program code.

[0187] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0188] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple flow charts and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0189] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple flow charts and / or boxes Figure 1 The function specified in one or more boxes.

[0190] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple flow charts and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0191] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction comprises one or more executable instructions for realizing the logical function of the specification. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the function or action of the specification, or can be implemented by a combination of dedicated hardware and computer instructions.

[0192] Finally, it should be noted that the embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which are all protected by the present invention.

Claims

1. A method for tunnel inspection using a drone, characterized in that: The method comprises: Obtain tunnel point cloud data of the tunnel to be inspected; Determining the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data; Based on the tunnel skeleton structure, the inspection path information of the tunnel to be inspected is determined, so that the drone inspects the tunnel to be inspected based on the inspection path information and generates tunnel inspection information.

2. The UAV tunnel inspection method according to claim 1, characterized in that: The determining of the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data includes: Determining the point cloud density of each point based on the tunnel point cloud data; According to the point cloud density of each point, each point is screened to obtain a candidate skeleton point; the candidate skeleton point is a point whose point cloud density is greater than or equal to a preset density threshold; Based on the candidate skeleton points and in combination with a preset skeleton extraction algorithm, the tunnel skeleton structure is determined.

3. The UAV tunnel inspection method according to claim 1, characterized in that: The determining, based on the tunnel skeleton structure, inspection path information of the tunnel to be inspected includes: vertically slicing the tunnel skeleton structure along the skeleton line to determine a plurality of skeleton cutting points; Determining multiple inspection points of the tunnel to be inspected according to the multiple skeleton tangent points; The inspection path information is determined based on the multiple inspection points.

4. The UAV tunnel inspection method according to claim 3, characterized in that: The step of determining a plurality of inspection points of the tunnel to be inspected based on the plurality of skeleton tangent points includes: Taking each skeleton cut point as the center and combining it with the sampling radius, multiple sampling areas are determined; According to each sampling area, determining a set of sampling points corresponding to each sampling area in the tunnel point cloud data; The sampling points in each sampling point set are moved along a target direction by a preset distance to obtain a plurality of moved sampling points, and the plurality of moved sampling points are used as the plurality of inspection points.

5. The UAV tunnel inspection method according to claim 3, characterized in that: The determining the inspection path information based on the multiple inspection points includes: Based on the multiple inspection points, construct an inspection path determination problem; Based on a preset path determination algorithm, the inspection path determination problem is solved to determine the inspection path information.

6. The UAV tunnel inspection method according to claim 5, characterized in that: The problem of constructing an inspection path determination problem based on the multiple inspection points includes: Determining the plurality of inspection points as a plurality of vertices of a path graph to be determined; Connecting each vertex to obtain multiple edges of the path graph to be determined; Determine the edge weight corresponding to each edge based on the distance between each vertex; Constructing the to-be-determined path graph according to a plurality of vertices of the to-be-determined path graph, a plurality of edges of the to-be-determined path graph, and an edge weight corresponding to each edge; Based on the path graph to be determined, the inspection path determination problem is constructed; the inspection path determination problem is to determine the shortest path traversing to each vertex of the path graph to be determined; wherein, the constraint conditions of the inspection path determination problem are that each vertex of the path graph to be determined is visited only once; and the path that can traverse to each vertex of the path graph to be determined has no self-intersection.

7. The UAV tunnel inspection method according to claim 6, characterized in that: Solving the inspection path determination problem based on a preset path determination algorithm and determining the inspection path information includes: Setting a trace strength for each edge of the path graph to be determined; wherein the trace strength is reduced proportionally; Set a preset number of path exploration bodies; Randomly selecting a vertex from among the multiple vertices of the to-be-determined path graph as a starting point through each path exploration body; and determining a current path exploration point among a plurality of vertices of the to-be-determined path graph based on the trace strength; and determining a current exploration path based on the current path exploration point; and in the process of the path exploration body exploring the current exploration path, updating the trace strength corresponding to the current exploration path based on the edge weight of the current exploration path, jumping to the step of determining the current path exploration point among the multiple vertices of the path graph to be determined based on the trace strength, until the path exploration body completes the exploration of the multiple vertices of the path graph to be determined; determining the exploration completion path of the path exploration body, and updating the path trace strength of the exploration completion path; Returning to each path exploration volume, randomly selecting a vertex from the multiple vertices of the path graph to be determined as a starting point until the number of iterations reaches a preset number; The exploration completed path whose path trace intensity is greater than a preset intensity threshold is determined as the inspection path information.

8. A UAV tunnel inspection device, characterized in that: The device comprises: A point cloud data acquisition module is used to acquire tunnel point cloud data of the tunnel to be inspected; A skeleton structure determination module, configured to determine the tunnel skeleton structure of the tunnel to be inspected based on the tunnel point cloud data; The inspection path determination module is used to determine the inspection path information of the tunnel to be inspected based on the tunnel skeleton structure, so that the drone can inspect the tunnel to be inspected based on the inspection path information and generate tunnel inspection information.

9. An electronic device comprising a processor and a memory, wherein the memory stores at least one instruction and at least one program, and the at least one instruction and the at least one program are loaded and executed by the processor to implement the drone tunnel inspection method as described in any one of claims 1 to 7.

10. A computer storage medium, wherein at least one instruction and at least one program are stored in the computer storage medium, and the at least one instruction and the at least one program are loaded and executed by a processor to implement the drone tunnel inspection method according to any one of claims 1 to 7.

Images