A route planning method and device based on subway tunnel inspection

Abstract

The present invention relates to the technical field of route planning, and discloses a route planning method and device based on subway tunnel inspection, the method comprising: performing relative positioning fusion on multi-source point cloud data of a target subway tunnel to obtain fused point cloud data; performing intensity correction on the fused point cloud data to obtain corrected point cloud data, and generating a surface grayscale map according to the corrected point cloud data; constructing a subway tunnel model according to the fused point cloud data, and identifying tunnel inspection targets according to the surface grayscale map; generating a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model, performing feasibility screening on the target map to obtain a feasible map; converting the feasible map into a rasterized map, performing optimal path planning according to the rasterized map, and obtaining a target inspection route of the target subway tunnel. The present invention can improve the planning efficiency and accuracy of subway tunnel inspection routes.

Description

Technical Field

[0001] The present invention relates to the technical field of route planning, and in particular to a route planning method and device based on subway tunnel inspection. Background Art

[0002] Subway tunnels are an important part of the subway system. The stability and safety of their structures are directly related to the safety of passengers and the reliability of subway operations. Therefore, it is crucial to conduct regular or irregular inspections and maintenance of subway tunnels to ensure the safety of subway operations and the integrity of tunnel structures. With the development of technology, subway tunnel inspections are increasingly using automated and intelligent equipment, such as robots, drones, sensor networks, etc. These devices can improve the efficiency and accuracy of inspections and reduce the risks of manual inspections.

[0003] In subway tunnel inspection, path planning is a key link to ensure the effectiveness and safety of subway tunnel inspection work. Through intelligent path planning, the efficiency and quality of inspection can be improved, safety risks can be reduced, and a scientific basis can be provided for tunnel maintenance and management. However, the existing inspection route planning methods mainly use the Ant Colony Optimization algorithm or the improved moth-to-flame algorithm to complete the inspection path planning. However, the large computational cost cannot meet the real-time planning requirements of the inspection path and the planning effect is not ideal in complex situations. Therefore, how to improve the planning efficiency and accuracy of subway tunnel inspection routes has become an urgent problem to be solved. Summary of the invention

[0004] The present invention provides a route planning method and device based on subway tunnel inspection, the main purpose of which is to solve the problem of poor planning efficiency and accuracy of subway tunnel inspection routes.

[0005] To achieve the above object, the present invention provides a route planning method based on subway tunnel inspection, comprising:

[0006] Acquire multi-source point cloud data of a target subway tunnel, and perform relative positioning fusion on the multi-source point cloud data to obtain fused point cloud data;

[0007] Performing intensity correction on the fused point cloud data to obtain corrected point cloud data, and generating a surface grayscale image of the target subway tunnel according to the corrected point cloud data;

[0008] constructing a subway tunnel model of the target subway tunnel according to the fused point cloud data, and identifying a tunnel inspection target of the target subway tunnel according to the surface grayscale image;

[0009] Generate a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model, and perform feasibility screening on the target map to obtain a feasible map;

[0010] The feasible map is converted into a rasterized map, and optimal path planning is performed according to the rasterized map to obtain a target inspection route for the target subway tunnel.

[0011] Optionally, performing relative positioning fusion on the multi-source point cloud data to obtain fused point cloud data includes:

[0012] Selecting reference kilometer mark data from the rotation speed sensor data set in the multi-source point cloud data;

[0013] Perform kilometer mark alignment according to the reference kilometer mark data and the segment data of the laser radar data set in the multi-source point cloud data to obtain an aligned kilometer mark;

[0014] Divide the laser displacement sensor data set in the multi-source point cloud data into intervals to obtain fastener interval kilometer markers;

[0015] Using the fastener interval kilometer mark to calibrate and locate the alignment kilometer mark to obtain a fused kilometer mark;

[0016] The alignment kilometer mark is calibrated and positioned using the following formula to obtain a fused kilometer mark: ;

[0017] Among them, r represents the fastener interval of the interval division, Indicates the fused kilometer mark corresponding to the kilometer mark of the fastener interval in the current fastener interval. Indicates the alignment kilometer mark of the current fastener interval. Indicates the current fastener interval. Fastener interval kilometer mark, Indicates the current fastener interval. Align the kilometer markers, Indicates the kilometer mark of the Eth fastener interval of the current fastener interval. The current fastener interval E

[0018] Align the kilometer markers;

[0019] The laser radar data set is calibrated according to the fused kilometer marker to obtain fused point cloud data.

[0020] Optionally, aligning the kilometer markers according to the reference kilometer marker data and the segment data of the laser radar data set in the multi-source point cloud data to obtain the aligned kilometer markers includes:

[0021] Calculate the kilometer mark scaling ratio of the segment data according to the reference kilometer mark data;

[0022] The kilometer scale ratio is calculated using the following formula: ;

[0023] in, Indicates the kilometer scale. Represents the section cross section, Indicates the kilometer mark data corresponding to the mth section in the benchmark kilometer mark data. Indicates the first kilometer in the benchmark data The kilometer mark data corresponding to each section section, Indicates the kilometer mark data corresponding to the mth section in the segment data. Indicates the first The kilometer mark data corresponding to each section section;

[0024] Perform kilometer mark alignment on the segment data according to the kilometer mark scaling ratio to obtain an aligned kilometer mark;

[0025] The segment data is aligned to the kilometer mark using the following formula to obtain the aligned kilometer mark: ;

[0026] in, Represents the section cross section, Indicates the alignment kilometer mark corresponding to the section in the segment data. Indicates the first kilometer in the benchmark data The kilometer mark data corresponding to each section section, Indicates the kilometer scale. Indicates the kilometer mark data corresponding to the section in the segment data. Indicates the first The kilometer mark data corresponding to each section.

[0027] Optionally, performing intensity correction on the fused point cloud data to obtain corrected point cloud data includes:

[0028] Calculate the distance between measuring points according to the three-dimensional coordinates in the fused point cloud data, and fit a distance polynomial according to the distance between measuring points;

[0029] Constructing a correction function for the fused point cloud data according to the distance polynomial;

[0030] The correction function is shown below: ;

[0031] Wherein, G represents the correction function, n represents the measurement point distance of the nth fused point cloud data, and N represents the total number of the fused point cloud data. represents the nth coefficient in the distance polynomial, and D represents the distance of the measuring point;

[0032] The reflection intensity of the fused point cloud is corrected according to the correction function to obtain corrected point cloud data.

[0033] Optionally, generating a surface grayscale image of the target subway tunnel according to the corrected point cloud data includes:

[0034] Mapping the three-dimensional coordinates of the corrected point cloud data into a preset two-dimensional plane image to obtain two-dimensional point cloud coordinates;

[0035] Performing intensity mapping according to the corrected point cloud data to obtain pixel values ​​corresponding to the coordinates of the two-dimensional point cloud, and normalizing the pixel values ​​to obtain coordinate grayscale values;

[0036] A surface grayscale image of the target subway tunnel is generated according to the two-dimensional point cloud coordinates and the coordinate grayscale values.

[0037] Optionally, constructing a subway tunnel model of the target subway tunnel according to the fused point cloud data includes:

[0038] Performing point cloud data filtering on the fused point cloud data to obtain target point cloud data;

[0039] Performing data mapping on the target point cloud data to obtain mapped point cloud data;

[0040] Three-dimensional modeling is performed according to the mapped point cloud data to obtain a subway tunnel model of the target subway tunnel.

[0041] Optionally, the identifying the tunnel inspection target of the target subway tunnel according to the surface grayscale image includes:

[0042] Using a preset convolution layer to perform convolution processing on the surface grayscale image, to obtain convolution features corresponding to a plurality of different convolution layers;

[0043] Performing convolution and upsampling processing on the convolution features respectively to obtain upsampled features;

[0044] Performing feature fusion on the up-sampled features to obtain target fusion features of the surface grayscale image;

[0045] The up-sampled features are fused using the following formula: ;

[0046] in, express A collection of convolution, batch normalization, and activation function processing, represents the learning weight corresponding to the upsampled features of the kth layer, represents the upsampled features of the kth layer, Indicates The learning weights corresponding to the upsampled features of the layer, represents the upsampled features of the kth layer, K represents the total number of convolutional layers, The splicing process is performed along the channel direction, and y represents the target fusion feature;

[0047] A tunnel inspection target of the target subway tunnel is identified according to the target fusion feature.

[0048] Optionally, generating a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model includes:

[0049] Determining the terrain elements of the target subway tunnel according to the subway tunnel model;

[0050] Constructing a plane map of the target subway tunnel according to the terrain elements;

[0051] The plane map is marked with targets using the tunnel inspection targets to obtain a target map of the target subway tunnel.

[0052] Optionally, performing optimal path planning according to the rasterized map to obtain a target inspection route of the target subway tunnel includes:

[0053] Determining inspection target points of the rasterized map;

[0054] Initialize a particle population, traverse the inspection target points according to the particle population, and obtain a planning path set corresponding to the particle population;

[0055] The particle population is optimized according to the path length corresponding to each planned path in the planned path set to obtain a target inspection route for the target subway tunnel.

[0056] In order to solve the above problems, the present invention also provides a route planning device based on subway tunnel inspection, the device comprising:

[0057] A relative positioning fusion module is used to obtain multi-source point cloud data of a target subway tunnel, and to perform relative positioning fusion on the multi-source point cloud data to obtain fused point cloud data;

[0058] A surface grayscale image generating module, used for performing intensity correction on the fused point cloud data to obtain corrected point cloud data, and generating a surface grayscale image of the target subway tunnel according to the corrected point cloud data;

[0059] A subway tunnel model and tunnel inspection target generation module, used to construct a subway tunnel model of the target subway tunnel according to the fused point cloud data, and to identify the tunnel inspection target of the target subway tunnel according to the surface grayscale image;

[0060] A feasible map construction module is used to generate a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model, and perform feasibility screening on the target map to obtain a feasible map;

[0061] The optimal path planning module is used to convert the feasible map into a rasterized map, perform optimal path planning according to the rasterized map, and obtain the target inspection route of the target subway tunnel.

[0062] The embodiment of the present invention can reduce the positioning error of three-dimensional point cloud data and obtain more accurate fused point cloud data by relatively positioning and fusing multi-source point cloud data of the target subway tunnel; perform intensity correction on the fused point cloud data to more accurately display the real situation of the target subway tunnel surface, and then generate a more accurate surface grayscale map; construct a subway tunnel model based on the fused point cloud data, and identify the tunnel inspection target based on the surface grayscale map, so as to generate a feasible map of the target subway tunnel, comprehensively display the feasible routes and structures of the target subway tunnel, and improve the accuracy and pertinence of subsequent optimal route planning; then convert the feasible map into a rasterized map, and divide the feasible map into grids of the same size to perform fast and accurate optimal path planning, effectively improving the planning efficiency and accuracy of subway tunnel inspection routes. Therefore, the route planning method and device based on subway tunnel inspection proposed by the present invention can solve the problem of poor planning efficiency and accuracy of subway tunnel inspection routes. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic flow chart of a route planning method based on subway tunnel inspection provided by an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of a process for performing intensity correction on fused point cloud data provided by an embodiment of the present invention;

[0065] Figure 3 A schematic diagram of a process of constructing a subway tunnel model of a target subway tunnel according to fused point cloud data provided by an embodiment of the present invention;

[0066] Figure 4 A functional module diagram of a route planning device based on subway tunnel inspection provided by one embodiment of the present invention.

[0067] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0068] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0069] The embodiment of the present application provides a route planning method based on subway tunnel inspection. The execution subject of the route planning method based on subway tunnel inspection includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the route planning method based on subway tunnel inspection can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.

[0070] Reference Figure 1 FIG. 1 is a flow chart of a route planning method based on subway tunnel inspection provided by an embodiment of the present invention. In this embodiment, the route planning method based on subway tunnel inspection includes:

[0071] S1. Acquire multi-source point cloud data of a target subway tunnel, and perform relative positioning fusion on the multi-source point cloud data to obtain fused point cloud data.

[0072] In the embodiment of the present invention, the target subway tunnel is a subway tunnel that needs to be inspected, wherein the subway tunnel connects different areas of the city and can be circular, rectangular, arched, etc. Multi-source point cloud data is the sensor data and different feature information of the subway tunnel line obtained by multiple sensors, such as a speed sensor, a laser radar, and a laser displacement sensor, wherein the speed sensor can obtain the kilometer mark of the subway tunnel by recording the number of pulses, the laser radar can obtain the section information of the tunnel body, platform, etc., and the laser displacement sensor can obtain the information of the fasteners and track plates on the subway tunnel.

[0073] In the embodiment of the present invention, the relative positioning fusion of the multi-source point cloud data to obtain fused point cloud data includes:

[0074] Selecting reference kilometer mark data from the rotation speed sensor data set in the multi-source point cloud data;

[0075] Perform kilometer mark alignment according to the reference kilometer mark data and the segment data of the laser radar data set in the multi-source point cloud data to obtain an aligned kilometer mark;

[0076] Divide the laser displacement sensor data set in the multi-source point cloud data into intervals to obtain fastener interval kilometer markers;

[0077] Using the fastener interval kilometer mark to calibrate and locate the alignment kilometer mark to obtain a fused kilometer mark;

[0078] The laser radar data set is calibrated according to the fused kilometer marker to obtain fused point cloud data.

[0079] In an embodiment of the present invention, the multi-source point cloud data includes data sets collected by a rotation speed sensor, a laser radar, and a laser displacement sensor, wherein the rotation speed sensor obtains the kilometer marker by recording the number of pulses to obtain a kilometer marker data set, i.e., the marker of each full kilometer of the target subway tunnel starting from the starting point. Multiple kilometer marker data with the smallest error with the preset subway data can be selected from the kilometer marker data set as benchmark kilometer marker data.

[0080] In detail, the LiDAR data set measures the distance data of objects by emitting laser beams and receiving reflected light. Since the inner wall shapes of the platform section and the tunnel section are quite different, the LiDAR data can be used to divide the target subway tunnel into different sections and the kilometer markers corresponding to the sections of different sections, so that the errors within the sections can be calibrated.

[0081] Specifically, the step of aligning the kilometer markers according to the reference kilometer marker data and the segment data of the laser radar data set in the multi-source point cloud data to obtain the aligned kilometer markers includes:

[0082] Calculate the kilometer mark scaling ratio of the segment data according to the reference kilometer mark data;

[0083] The segment data is aligned with kilometer marks according to the kilometer mark scaling ratio to obtain an aligned kilometer mark.

[0084] Specifically, the kilometer scale ratio is calculated using the following formula: ;

[0085] in, Indicates the kilometer scale. Represents the section cross section, Indicates the kilometer mark data corresponding to the mth section in the benchmark kilometer mark data. Indicates the first kilometer in the benchmark data The kilometer mark data corresponding to each section section, Indicates the kilometer mark data corresponding to the mth section in the segment data. Indicates the first The kilometer mark data corresponding to each section.

[0086] Furthermore, the segment data is aligned with the kilometer mark using the following formula to obtain the aligned kilometer mark: ;

[0087] in, Represents the section cross section, Indicates the alignment kilometer mark corresponding to the section in the segment data. Indicates the first kilometer in the benchmark data The kilometer mark data corresponding to each section section, Indicates the kilometer scale. Indicates the kilometer mark data corresponding to the section in the segment data. Indicates the first The kilometer mark data corresponding to each section.

[0088] In an embodiment of the present invention, the target subway tunnel can be divided into more precise fastener sections through the laser displacement sensor data set, and a certain section of the target subway tunnel can be divided into different fastener intervals, thereby obtaining the fastener interval kilometer markers, and the fastener interval kilometer markers can be used to perform more precise calibration and positioning of the alignment kilometer markers, thereby obtaining the fusion positioning points of the fusion speed sensor, lidar and laser displacement sensor.

[0089] Specifically, the alignment kilometer mark is calibrated and positioned using the following formula to obtain a fused kilometer mark: ;

[0090] Among them, r represents the fastener interval of the interval division, Indicates the fused kilometer mark corresponding to the kilometer mark of the fastener interval in the current fastener interval. Indicates the alignment kilometer mark of the current fastener interval. Indicates the current fastener interval. Fastener interval kilometer mark, Indicates the current fastener interval. Align the kilometer markers, Indicates the kilometer mark of the Eth fastener interval of the current fastener interval. The Eth aligned kilometer marker of the current fastener interval.

[0091] In an embodiment of the present invention, by fusing kilometer markers to finely calibrate the scanned kilometer markers in the laser radar data set, the positioning error of the three-dimensional point cloud data in the laser radar data set can be reduced to obtain more accurate fused point cloud data.

[0092] S2. Performing intensity correction on the fused point cloud data to obtain corrected point cloud data, and generating a surface grayscale image of the target subway tunnel according to the corrected point cloud data.

[0093] In the embodiment of the present invention, the three-dimensional point cloud data acquired by the laser radar is affected by multiple variables, among which the incident angle and distance play a key role. Considering that the influence of distance on the corrected intensity value is only affected by the reflectivity of the object surface, the corrected laser reflection intensity value can more accurately show the real situation of the target subway tunnel surface. In detail, the fused point cloud data contains the three-dimensional coordinates and laser reflection intensity (Intensity) of each point. The laser reflection intensity reflects the intensity of the laser signal reflected back after the radar laser beam hits the target subway tunnel surface.

[0094] In the embodiment of the present invention, participation Figure 2 As shown, the intensity correction of the fused point cloud data to obtain the corrected point cloud data includes:

[0095] S21, calculating the distance of the measuring points according to the three-dimensional coordinates in the fused point cloud data, and fitting a distance polynomial according to the distance of the measuring points;

[0096] S22, constructing a correction function for the fused point cloud data according to the distance polynomial;

[0097] S23. Correct the reflection intensity of the fused point cloud according to the correction function to obtain corrected point cloud data.

[0098] In an embodiment of the present invention, the measuring point distance of each fused point cloud data can be calculated based on the three-dimensional coordinates and the horizontal and vertical scanning angles calculated by the lidar scanning, and then a polynomial of a preset order can be fitted. For example, an n-order linear equation group can be constructed based on the measuring point distances of n+1 fused point cloud data, and the n-order linear equation group can be solved to obtain the distance polynomial.

[0099] In detail, the correction function is shown as follows: ;

[0100] Wherein, G represents the correction function, n represents the measurement point distance of the nth fused point cloud data, and N represents the total number of the fused point cloud data. represents the nth coefficient in the distance polynomial, and D represents the distance of the measuring point.

[0101] In the embodiment of the present invention, the correction function is multiplied by the laser reflection intensity of each fused point cloud data to obtain the corrected point cloud data, which can more accurately show the real situation of the target subway tunnel surface.

[0102] Furthermore, the surface grayscale map is a grayscale image of a two-dimensional plane converted from the corrected point cloud data, which is beneficial for identifying the topographic observation of the target subway tunnel and facilitating the subsequent construction of a subway tunnel model of the target subway tunnel.

[0103] In detail, generating the surface grayscale image of the target subway tunnel according to the corrected point cloud data includes:

[0104] Mapping the three-dimensional coordinates of the corrected point cloud data into a preset two-dimensional plane image to obtain two-dimensional point cloud coordinates;

[0105] Performing intensity mapping according to the corrected point cloud data to obtain pixel values ​​corresponding to the coordinates of the two-dimensional point cloud, and normalizing the pixel values ​​to obtain coordinate grayscale values;

[0106] A surface grayscale image of the target subway tunnel is generated according to the two-dimensional point cloud coordinates and the coordinate grayscale values.

[0107] In an embodiment of the present invention, the three-dimensional coordinates can be projected onto the two-dimensional image plane according to the internal parameters of the laser radar (for example, the focal length and the coordinates of the principal point, etc.) to obtain the two-dimensional point cloud coordinates, and the reflection intensity value corresponding to each corrected point cloud data is mapped to the corresponding pixel position of the two-dimensional image to obtain the pixel value corresponding to each two-dimensional point cloud coordinate. The generated pixel value range is usually 0 to 255, which reflects the strength of the laser reflection signal. The higher the intensity value, the brighter the grayscale. The coordinate grayscale value is then obtained through normalization processing.

[0108] Furthermore, a blank two-dimensional array is created, whose number of rows and columns corresponds to the length and width of the image. Then, the normalized coordinate grayscale values ​​are traversed, and the coordinate grayscale value of each two-dimensional point cloud coordinate is filled into the corresponding pixel position to obtain the surface grayscale image of the target subway tunnel.

[0109] S3. Constructing a subway tunnel model of the target subway tunnel according to the fused point cloud data, and identifying a tunnel inspection target of the target subway tunnel according to the surface grayscale image.

[0110] In the embodiment of the present invention, the subway tunnel model is a three-dimensional model for comprehensively displaying the distribution and internal structure of various equipment in the target subway tunnel.

[0111] In the embodiment of the present invention, refer to Figure 3 As shown, the step of constructing a subway tunnel model of the target subway tunnel according to the fused point cloud data includes:

[0112] S31, performing point cloud data filtering on the fused point cloud data to obtain target point cloud data;

[0113] S32, performing data mapping on the target point cloud data to obtain mapped point cloud data;

[0114] S33. Perform three-dimensional modeling according to the mapped point cloud data to obtain a subway tunnel model of the target subway tunnel.

[0115] In an embodiment of the present invention, point cloud data filtering can remove noise points, outliers and unnecessary data in the fused point cloud data, thereby improving the quality of the fused point cloud data and the accuracy of subsequent processing. Specifically, point cloud data filtering can be performed by filtering methods such as Gaussian filtering or bilateral filtering.

[0116] Furthermore, the RBF (radial basis function) neural network can be used to interpolate the fused point cloud data to generate mapped point cloud data, where the RBF neural network consists of an input layer, a hidden layer, and an output layer. The input layer is used to receive the target point cloud data, and the hidden layer is composed of radial basis function neurons. Each neuron corresponds to a radial basis function, usually a Gaussian function. These functions map the target point cloud data and map the target point cloud data space to a high-dimensional space, so that the originally linearly inseparable target point cloud data may become linearly separable in the new space. The output layer is used to output the mapped point cloud data in the high-dimensional space.

[0117] In detail, the target subway tunnel can be three-dimensionally modeled by mapping the point cloud data, for example, using BIM (Building Information Modeling) technology to comprehensively display the distribution and internal structure of various equipment in the target subway tunnel.

[0118] In an embodiment of the present invention, the step of identifying the tunnel inspection target of the target subway tunnel according to the surface grayscale image includes:

[0119] Using a preset convolution layer to perform convolution processing on the surface grayscale image, to obtain convolution features corresponding to a plurality of different convolution layers;

[0120] Performing convolution and upsampling processing on the convolution features respectively to obtain upsampled features;

[0121] Performing feature fusion on the up-sampled features to obtain target fusion features of the surface grayscale image;

[0122] A tunnel inspection target of the target subway tunnel is identified according to the target fusion feature.

[0123] In an embodiment of the present invention, a pre-trained convolutional model can be used to identify tunnel inspection targets. For example, a surface grayscale image is convolved using multiple 3*3 convolutional layers to obtain convolution features corresponding to multiple different convolutional layers, and then a 3*3 convolutional layer is used for convolution and upsampling is performed at the same time to obtain upsampling features.

[0124] Specifically, the up-sampled features are fused using the following formula: ;

[0125] in, express A collection of convolution, batch normalization, and activation function processing, represents the learning weight corresponding to the upsampled features of the kth layer, represents the upsampled features of the kth layer, Indicates The learning weights corresponding to the upsampled features of the layer, represents the upsampled features of the kth layer, K represents the total number of convolutional layers, The splicing process is performed along the channel direction, and y represents the target fusion feature.

[0126] In an embodiment of the present invention, the target subway tunnel may contain hidden danger features such as water leakage, pipelines, corrosion, and tunnel inspection targets that need to be inspected in depth, such as tunnel structures. By matching the target fusion features with preset standard features, the tunnel inspection targets in the target subway tunnel can be identified, thereby improving the accuracy of path planning.

[0127] S4. Generate a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model, and perform feasibility screening on the target map to obtain a feasible map.

[0128] In an embodiment of the present invention, the target map is a map containing tunnel inspection targets and showing information such as the target subway tunnel structure, tunnel direction, stations, equipment, etc. The target map can be used to better understand the inspection route and improve the accuracy of subsequent target inspection route generation.

[0129] In an embodiment of the present invention, generating a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model includes:

[0130] Determining the terrain elements of the target subway tunnel according to the subway tunnel model;

[0131] Constructing a plane map of the target subway tunnel according to the terrain elements;

[0132] The plane map is marked with targets using the tunnel inspection targets to obtain a target map of the target subway tunnel.

[0133] In the embodiment of the present invention, the terrain element is an element that can represent the basic information of the target map, for example, the scale, the structures included in the target railway tunnel (tunnel, platform, fasteners and track plates, etc.), the boundary line of the target subway tunnel, the precise coordinates, the line of the target railway tunnel, etc. The three-dimensional subway tunnel model can be flattened to construct a plane map showing the information of the target subway tunnel, and the tunnel inspection targets can be marked at the corresponding positions on the plane map to obtain a target map containing the tunnel inspection targets.

[0134] Furthermore, feasibility screening is to screen out terrain in the target map that is impassable during the inspection process. For example, when using a robot for inspection, it is necessary to determine obstacles and fault areas in the target subway tunnel that hinder the robot's progress, thereby improving the accuracy of path planning.

[0135] In the embodiment of the present invention, the impassable tunnel inspection targets marked in the target map can be identified, and the target areas corresponding to the impassable tunnel inspection targets are marked as impassable in the target map to obtain a feasible map.

[0136] In the embodiment of the present invention, by generating a target map of a target subway tunnel and performing feasibility screening on the target map, the feasible routes and structures of the target subway tunnel can be fully displayed, thereby improving the accuracy of subsequent optimal route planning.

[0137] S5. Convert the feasible map into a rasterized map, perform optimal path planning according to the rasterized map, and obtain a target inspection route for the target subway tunnel.

[0138] In an embodiment of the present invention, a rasterized map is a map representation method that divides a map area into uniform grid units, and each grid unit can be assigned specific attribute information. For example, a feasible map is divided into grids of the same size, where inaccessible areas are filled with black and accessible areas are filled with white, thereby obtaining a rasterized map.

[0139] In the embodiment of the present invention, the optimal path planning is a path with the shortest distance to complete the inspection target of the target subway tunnel, which can improve the efficiency of the inspection of the target subway tunnel and ensure the accuracy of the inspection.

[0140] Specifically, performing optimal path planning according to the rasterized map to obtain a target inspection route of the target subway tunnel includes:

[0141] Determining inspection target points of the rasterized map;

[0142] Initialize a particle population, traverse the inspection target points according to the particle population, and obtain a planning path set corresponding to the particle population;

[0143] The particle population is optimized according to the path length corresponding to each planned path in the planned path set to obtain a target inspection route for the target subway tunnel.

[0144] In an embodiment of the present invention, the inspection target points are the positions of the tunnel inspection targets that need to be inspected in a rasterized map and the starting and ending points of the rasterized map. By traversing the inspection target points, it can be ensured that the path inspection can cover all target subway tunnels, ensure the accuracy and pertinence of the target inspection route, and improve the planning efficiency of the subway tunnel inspection route.

[0145] In detail, the particle population is the set of all individuals (or chromosomes) in each generation of the genetic algorithm. Each individual represents a planning path, which transforms the solution process of the target inspection route into an evolutionary solution process. Through the continuous iteration of inheritance, mutation, natural selection and hybridization of the particle population, the feasible solution gradually converges to the goal of minimizing the path length, and obtains the target inspection route that meets the requirements.

[0146] In the embodiment of the present invention, the target inspection route can be used to minimize the route distance of the target inspection route while ensuring the comprehensiveness and pertinence of the inspection, thereby improving the planning efficiency and accuracy of the subway tunnel inspection route.

[0147] like Figure 4 , which is a functional module diagram of a route planning device based on subway tunnel inspection provided by an embodiment of the present invention.

[0148] The route planning device 400 based on subway tunnel inspection of the present invention can be installed in an electronic device. According to the functions implemented, the route planning device 400 based on subway tunnel inspection can include a relative positioning fusion module 401, a surface grayscale image generation module 402, a subway tunnel model and tunnel inspection target generation module 403, a feasible map construction module 404 and an optimal path planning module 405. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0149] In this embodiment, the functions of each module / unit are as follows:

[0150] The relative positioning fusion module 401 is used to obtain multi-source point cloud data of a target subway tunnel, and perform relative positioning fusion on the multi-source point cloud data to obtain fused point cloud data;

[0151] The surface grayscale image generating module 402 is used to perform intensity correction on the fused point cloud data to obtain corrected point cloud data, and to generate a surface grayscale image of the target subway tunnel according to the corrected point cloud data;

[0152] The subway tunnel model and tunnel inspection target generation module 403 is used to construct a subway tunnel model of the target subway tunnel according to the fused point cloud data, and to identify the tunnel inspection target of the target subway tunnel according to the surface grayscale image;

[0153] The feasible map construction module 404 is used to generate a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model, and perform feasibility screening on the target map to obtain a feasible map;

[0154] The optimal path planning module 405 is used to convert the feasible map into a rasterized map, perform optimal path planning according to the rasterized map, and obtain a target inspection route for the target subway tunnel.

[0155] In detail, each module described in the route planning device 400 based on subway tunnel inspection in the embodiment of the present invention is used in the same manner as described above. Figures 1 to 3 The route planning method based on subway tunnel inspection described in the text is the same as the technical means and can produce the same technical effects, so I will not go into details here.

[0156] The present invention also provides an electronic device, which may include a processor, a memory, a communication bus and a communication interface, and may also include a computer program stored in the memory and executable on the processor, such as a route planning method program based on subway tunnel inspection.

[0157] In some embodiments, the processor may be composed of an integrated circuit, for example, it may be composed of a single packaged integrated circuit, or it may be composed of multiple packaged integrated circuits with the same or different functions, including one or more central processing units (CPU), microprocessors, digital processing chips, graphics processors and a combination of various control chips.

[0158] The memory includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device.

[0159] The communication bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory and at least one processor, etc.

[0160] The communication interface is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is usually used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a display, an input unit (such as a keyboard), and optionally, the user interface may also be a standard wired interface or a wireless interface.

[0161] The figure only shows an electronic device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0162] Specifically, the specific implementation method of the processor for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0163] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0164] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0165] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0166] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0167] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0168] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application device that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0169] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in a device claim can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A route planning method based on subway tunnel inspection, characterized in that: The method comprises: Acquire multi-source point cloud data of a target subway tunnel, perform relative positioning fusion on the multi-source point cloud data, and obtain fused point cloud data; wherein, the relative positioning fusion on the multi-source point cloud data to obtain fused point cloud data includes: selecting reference kilometer mark data from the rotation speed sensor data set in the multi-source point cloud data; aligning kilometer marks according to the reference kilometer mark data and the segment data of the laser radar data set in the multi-source point cloud data to obtain aligned kilometer marks; dividing the laser displacement sensor data set in the multi-source point cloud data into intervals to obtain fastener interval kilometer marks; calibrating and positioning the alignment kilometer marks using the fastener interval kilometer marks to obtain fused kilometer marks; The alignment kilometer mark is calibrated and positioned using the following formula to obtain a fused kilometer mark: ; in, represents the fastener interval of the interval division, Indicates the fused kilometer mark corresponding to the kilometer mark of the fastener interval in the current fastener interval. Indicates the alignment kilometer mark of the current fastener interval. Indicates the current fastener interval. Fastener interval kilometer mark, Indicates the current fastener interval. Align the kilometer markers, Indicates the current fastener interval. Fastener interval kilometer mark, The current fastener interval Align the kilometer markers; Calibrate the laser radar data set according to the fused kilometer mark to obtain fused point cloud data; Performing intensity correction on the fused point cloud data to obtain corrected point cloud data, and generating a surface grayscale image of the target subway tunnel according to the corrected point cloud data; constructing a subway tunnel model of the target subway tunnel according to the fused point cloud data, and identifying a tunnel inspection target of the target subway tunnel according to the surface grayscale image; Generate a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model, and perform feasibility screening on the target map to obtain a feasible map; The feasible map is converted into a rasterized map, and optimal path planning is performed according to the rasterized map to obtain a target inspection route for the target subway tunnel.

2. The route planning method based on subway tunnel inspection according to claim 1, characterized in that: The step of aligning the kilometer markers according to the reference kilometer marker data and the segment data of the laser radar data set in the multi-source point cloud data to obtain the aligned kilometer markers includes: Calculate the kilometer mark scaling ratio of the segment data according to the reference kilometer mark data; The kilometer scale ratio is calculated using the following formula: ; in, Indicates the kilometer scale. Represents the section cross section, Indicates the first kilometer in the benchmark data The kilometer mark data corresponding to each section section, Indicates the first kilometer in the benchmark data The kilometer mark data corresponding to each section section, Indicates the first The kilometer mark data corresponding to each section section, Indicates the first The kilometer mark data corresponding to each section section; Perform kilometer mark alignment on the segment data according to the kilometer mark scaling ratio to obtain an aligned kilometer mark; The segment data is aligned to the kilometer mark using the following formula to obtain the aligned kilometer mark: ; in, Indicates the alignment kilometer mark corresponding to the section in the segment data. Represents the kilometer mark data corresponding to the section in the segment data.

3. The route planning method based on subway tunnel inspection according to claim 1, characterized in that: The step of performing intensity correction on the fused point cloud data to obtain corrected point cloud data includes: Calculate the distance between measuring points according to the three-dimensional coordinates in the fused point cloud data, and fit a distance polynomial according to the distance between measuring points; Constructing a correction function for the fused point cloud data according to the distance polynomial; The correction function is shown below: ; Wherein, G represents the correction function, n represents the nth fused point cloud data, and N represents the total number of fused point cloud data. represents the nth coefficient in the distance polynomial, Indicates the measurement point distance of the nth fused point cloud data; The reflection intensity of the fused point cloud is corrected according to the correction function to obtain corrected point cloud data.

4. The route planning method based on subway tunnel inspection according to claim 1, characterized in that: The step of generating a surface grayscale image of the target subway tunnel according to the corrected point cloud data comprises: Mapping the three-dimensional coordinates of the corrected point cloud data into a preset two-dimensional plane image to obtain two-dimensional point cloud coordinates; Performing intensity mapping according to the corrected point cloud data to obtain pixel values ​​corresponding to the coordinates of the two-dimensional point cloud, and normalizing the pixel values ​​to obtain coordinate grayscale values; A surface grayscale image of the target subway tunnel is generated according to the two-dimensional point cloud coordinates and the coordinate grayscale values.

5. The route planning method based on subway tunnel inspection according to claim 1, characterized in that: The step of constructing a subway tunnel model of the target subway tunnel according to the fused point cloud data comprises: Performing point cloud data filtering on the fused point cloud data to obtain target point cloud data; Performing data mapping on the target point cloud data to obtain mapped point cloud data; Three-dimensional modeling is performed according to the mapped point cloud data to obtain a subway tunnel model of the target subway tunnel.

6. The route planning method based on subway tunnel inspection according to claim 1, characterized in that: The step of identifying the tunnel inspection target of the target subway tunnel according to the surface grayscale image includes: Using a preset convolution layer to perform convolution processing on the surface grayscale image, to obtain convolution features corresponding to a plurality of different convolution layers; Performing convolution and upsampling processing on the convolution features respectively to obtain upsampled features; Performing feature fusion on the up-sampled features to obtain target fusion features of the surface grayscale image; The up-sampled features are fused using the following formula: ; in, express A collection of convolution, batch normalization, and activation function processing, represents the learning weight corresponding to the upsampled features of the kth layer, represents the upsampled features of the kth layer, Indicates The learning weights corresponding to the upsampled features of the layer, Indicates The upsampled features of the layer, K represents the total number of convolutional layers, The splicing process is performed along the channel direction, and y represents the target fusion feature; A tunnel inspection target of the target subway tunnel is identified according to the target fusion feature.

7. The route planning method based on subway tunnel inspection according to claim 1, characterized in that: The step of generating a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model includes: Determining the terrain elements of the target subway tunnel according to the subway tunnel model; Constructing a plane map of the target subway tunnel according to the terrain elements; The plane map is marked with targets using the tunnel inspection targets to obtain a target map of the target subway tunnel.

8. The route planning method based on subway tunnel inspection as claimed in claim 1, characterized in that: The optimal path planning is performed according to the rasterized map to obtain the target inspection route of the target subway tunnel, including: Determining inspection target points of the rasterized map; Initialize a particle population, traverse the inspection target points according to the particle population, and obtain a planning path set corresponding to the particle population; The particle population is optimized according to the path length corresponding to each planned path in the planned path set to obtain a target inspection route for the target subway tunnel.

9. A route planning device based on subway tunnel inspection, characterized in that: The device comprises: A relative positioning fusion module is used to obtain multi-source point cloud data of a target subway tunnel, and to perform relative positioning fusion on the multi-source point cloud data to obtain fused point cloud data; wherein, the relative positioning fusion on the multi-source point cloud data to obtain fused point cloud data includes: selecting reference kilometer mark data from the rotation speed sensor data set in the multi-source point cloud data; aligning kilometer marks according to the reference kilometer mark data and the segment data of the laser radar data set in the multi-source point cloud data to obtain aligned kilometer marks; dividing the laser displacement sensor data set in the multi-source point cloud data into intervals to obtain fastener interval kilometer marks; and calibrating and positioning the alignment kilometer marks using the fastener interval kilometer marks to obtain fused kilometer marks; The alignment kilometer mark is calibrated and positioned using the following formula to obtain a fused kilometer mark: ; Among them, r represents the fastener interval of the interval division, Indicates the fused kilometer mark corresponding to the kilometer mark of the fastener interval in the current fastener interval. Indicates the alignment kilometer mark of the current fastener interval. Indicates the current fastener interval. Fastener interval kilometer mark, Indicates the current fastener interval. Align the kilometer markers, Indicates the current fastener interval. Fastener interval kilometer mark, The current fastener interval Align the kilometer markers; Calibrate the laser radar data set according to the fused kilometer mark to obtain fused point cloud data; A surface grayscale image generating module, used for performing intensity correction on the fused point cloud data to obtain corrected point cloud data, and generating a surface grayscale image of the target subway tunnel according to the corrected point cloud data; A subway tunnel model and tunnel inspection target generation module, used to construct a subway tunnel model of the target subway tunnel according to the fused point cloud data, and to identify the tunnel inspection target of the target subway tunnel according to the surface grayscale image; A feasible map construction module is used to generate a target map of the target subway tunnel according to the tunnel inspection target and the subway tunnel model, and perform feasibility screening on the target map to obtain a feasible map; The optimal path planning module is used to convert the feasible map into a rasterized map, perform optimal path planning according to the rasterized map, and obtain the target inspection route of the target subway tunnel.

Images