Multi-sensor data fusion navigation method and system for cable tunnel robot

Abstract

The invention relates to the technical field of cable tunnel robot navigation, in particular to a cable tunnel robot multi-sensor data fusion navigation method and a cable tunnel robot multi-sensor data fusion navigation system. Extracting tunnel distribution characteristics, a tunnel entrance position, a tunnel exit position, task starting point position information, inspection task point position information and task end point position information from the inspection task data; generating an initial inspection path plan according to the inspection task data; environment data outside the tunnel robot and motion state data of the tunnel robot are collected; the environment data comprises one or more of distance data, thermal imaging data, video data and audio data; and tunnel environment modeling is carried out by utilizing external environment data of the tunnel robot, a state adjustment instruction is autonomously generated in combination with motion state data, and operation parameters of the tunnel robot are adjusted.

Description

Technical Field

[0001] The present invention relates to the technical field of cable tunnel robot navigation, and in particular to a cable tunnel robot multi-sensor data fusion navigation method and system. Background Art

[0002] With the increase of power load and the complexity of power system, tunnel robots are more and more widely used in power cable tunnels. As a typical complex environment, cable tunnels are characterized by narrow space, irregular terrain and possible obstacles. Quadruped robots are a type of tunnel robot. They have obvious advantages over wheeled and tracked robots in terms of obstacle crossing and passability. Therefore, they have better adaptability to complex terrains. When equipment fails, they can replace operation and maintenance personnel to approach the equipment, collect images and infrared information of the equipment, and help operation and maintenance personnel to grasp the status information of on-site equipment more timely and comprehensively, so as to make fault analysis and emergency disposal decisions faster and more accurately, limit the impact of faults to a minimum, and realize autonomous operation and maintenance of cable tunnels.

[0003] At present, traditional tunnel robot navigation technology has the following defects: 1. Insufficient environmental perception and unable to adapt to the complex cable tunnel environment; 2. Limitations of a single sensor; 3. The single sensor system has poor robustness to sensor failure or data noise. Summary of the invention

[0004] The present invention can enable the tunnel robot to perceive its surrounding environment, autonomously update the travel path, avoid obstacles, autonomously navigate in the cable tunnel environment, and cope with complex cable tunnel environments.

[0005] The technical solution proposed by the present invention is: a multi-sensor data fusion navigation method for a cable tunnel robot, the method comprising: Obtain inspection task data of the tunnel robot, and extract tunnel distribution characteristics, tunnel entrance location, tunnel exit location, task starting point location information, inspection task point location information, and task end point location information from the inspection task data; Generate initial inspection route planning based on inspection task data; Collecting environmental data outside the tunnel robot and motion state data of the tunnel robot, wherein the environmental data includes one or more of distance data, thermal imaging data, video data, and audio data; The tunnel environment is modeled using the robot's external environmental data, and combined with the motion state data, state adjustment instructions are autonomously generated to adjust the tunnel robot's operating parameters.

[0006] Preferably, extracting tunnel distribution characteristics from inspection task data comprises the following steps: Obtain the tunnel distribution map image, convert the map image into a grayscale image, and use high-speed blur to smooth the map image and reduce noise; Identify edges in map images using the Canny edge detection algorithm; Detect contours in map images using the OpenCV library; Perform feature extraction on the detected contour to obtain the area, perimeter, width, length and circularity of the contour; Set the contour area screening threshold to filter contours whose area is smaller than the screening threshold; Perform curve contour feature extraction on the remaining contours to obtain curve contour position features, length features and curvature features; Assigning identification labels to the extracted curve profiles; The width and length of the tunnel are determined using the identified contour width and length, and the angle characteristics of the tunnel curve are determined using the identified contour curvature characteristics.

[0007] Preferably, generating the initial inspection path planning according to the inspection task data comprises the following steps: Acquire path generation information, wherein the path generation information includes tunnel entrance location information, tunnel exit location information, tunnel length information, curve location information, and curve angle information; The path search algorithm A-star is used to generate preliminary inspection paths from the tunnel entrance and tunnel exit based on the acquired path generation information.

[0008] Preferably, the collecting of the external environment data of the tunnel robot and the motion state data of the tunnel robot comprises the following steps: Acquire point cloud data of the external environment of the tunnel robot through multi-line laser radar; Acquire thermal images of the external environment of the tunnel robot through an infrared thermal imaging device; Obtain high-definition images of the external environment of the tunnel robot through a high-definition camera; The distance data between the tunnel robot and the objects inside the tunnel is obtained through the ultrasonic ranging sensor; Obtain gas concentration data outside the tunnel robot through gas monitoring sensors; The motion state data of the tunnel robot is obtained through the inertial measurement unit and speed sensor.

[0009] Preferably, the tunnel environment modeling using the environment data outside the robot comprises the following steps: Read point cloud data, remove point cloud data noise, and then perform downsampling and filtering; Obtain tunnel distribution characteristics; Acquire infrared thermal images, perform denoising and contrast enhancement on the infrared thermal images; The point cloud data and the infrared imaging data are registered to align the point cloud data and the infrared imaging data in space; Use point cloud data to generate a 3D network model of the tunnel interior, and use infrared thermal imaging data to add texture information to the 3D model; Use Open3D library to visualize the 3D network model inside the tunnel; The method of generating a three-dimensional network model inside the tunnel using point cloud data comprises the following steps: Using a normal vector estimation algorithm, identifying key features inside the tunnel, including tunnel width, tunnel floor, and tunnel ceiling; The point cloud data is segmented by clustering algorithm or region growing algorithm, and the segmented point cloud data is classified by classifier to distinguish different key features; According to the classification results, the point cloud data inside the tunnel is reconstructed in 3D using the triangulation method to generate a 3D network model inside the tunnel.

[0010] Preferably, the tunnel environment modeling using the external environment data of the tunnel robot further includes the following steps: Obstacle point cloud data is obtained through multi-line laser radar; Use the density clustering DBSCAN algorithm to perform cluster analysis on obstacle point cloud data and identify different obstacles; Perform feature extraction on the obstacle point cloud data after cluster analysis to obtain the position, size, shape and volume features of different obstacles to form feature data sets of different obstacles; Importing the three-dimensional network model inside the tunnel, obtaining multiple node coordinates of the three-dimensional network model, traversing each node of the three-dimensional network model, and checking whether each node intersects with the obstacle to determine whether the obstacle intersects with the tunnel or is located inside the tunnel; According to the intersection position between the node and the obstacle, the obstacle information is added to the three-dimensional network model inside the tunnel, and the node status is updated, wherein the node status includes marking as requiring detour or impassable; Output the three-dimensional network model containing obstacle information.

[0011] Preferably, updating the initial inspection path planning according to the three-dimensional network model containing obstacle information comprises the following steps: Import a 3D network model and divide the 3D network model into multiple 3D grids; Set up a 3D array for each 3D grid ,in, , Represents the total number of three-dimensional grids; They represent the coordinates of the three-dimensional grid in the spatial coordinate system respectively; Traverse each 3D grid and check whether each 3D grid intersects with the obstacle. If it intersects with the obstacle, ,otherwise ; Based on the assignment of the three-dimensional array, the path search algorithm A-star is used to find the path from the starting point to the end point of the three-dimensional network model to update the initial inspection path; And send the updated inspection path to the tunnel robot.

[0012] Preferably, the step of combining the motion state data, autonomously generating a state adjustment instruction, and adjusting the operating parameters of the tunnel robot comprises the following steps: Get the current location information of the tunnel robot; Acquire current motion state data of the tunnel robot, wherein the motion state data includes motion speed, acceleration, angular velocity and posture of the tunnel robot; The tunnel robot The position information, motion speed, acceleration, angular velocity and posture data at the moment constitute the motion state feature set ,in Respectively The position vector, velocity vector, acceleration vector, angular velocity vector and attitude vector of the moment tunnel robot; , , , , ;in, represents the spatial coordinates of the tunnel robot, Tunnel robot Always Axis speed, The speed of the axis and Speed ​​in the axis direction; Tunnel robot Always The acceleration in the axial direction, The acceleration in the axial direction and Acceleration in the axis direction; Tunnel robot Always Angular velocity in the direction of the axis, The angular velocity of the axis and Angular velocity in the direction of the axis; Tunnel robot Roll angle, pitch angle and yaw angle at the moment; based on , through Kalman filter, predicting tunnel robot The state of motion at all times; Implemented through particle filter Data fusion in According to the predicted motion state, a tunnel robot control instruction is generated to control the tunnel robot to accelerate and / or turn.

[0013] The present invention also provides a cable tunnel robot multi-sensor data fusion navigation system, including a multi-sensor data acquisition module, a data fusion processing module, an autonomous navigation control module, a motion module and a feedback control module. The multi-sensor data acquisition module is used to obtain environmental data around the tunnel robot, the data fusion processing module performs fusion processing on the environmental data, the autonomous navigation control module makes path planning and motion instructions according to the fusion processing results, the motion module is used to control the motion state of the tunnel robot according to the path planning and motion instructions, the feedback control module is used to provide feedback on the motion instructions of the tunnel robot, and the system is used to execute the multi-sensor data fusion navigation method of the cable tunnel robot.

[0014] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the multi-sensor data fusion navigation method of a cable tunnel robot.

[0015] Beneficial effects of the present invention: 1. In the present invention, the tunnel robot can sense the external cable tunnel environment, obtain various environmental data through various sensors, generate an inspection path and a three-dimensional network model of the tunnel by analyzing the various environmental data, and optimize the inspection path based on the obstacle information if there are obstacles on the inspection path.

[0016] 2. In the present invention, the tunnel robot can obtain its own current motion state information, and predict the motion state at the next moment (specifically according to the acquisition frequency of the motion state) through the Kalman filter, and generate robot control instructions according to the predicted motion state to control the operation of the tunnel robot (acceleration, deceleration and steering), thereby ensuring the stable operation of the tunnel robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention provides a flow chart of a multi-sensor data fusion navigation method for a cable tunnel robot. DETAILED DESCRIPTION

[0018] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0019] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0020] refer to Figure 1 The technical solution provided by the present invention is: a multi-sensor data fusion navigation method for a cable tunnel robot, comprising the following steps: 1. Obtain inspection task data of the tunnel robot, and extract tunnel distribution characteristics, tunnel entrance position, tunnel exit position, task starting point position information, inspection task point position information, and task end point position information from the inspection task data; wherein, extracting tunnel distribution characteristics from the inspection task data includes the following steps: Obtain the tunnel distribution map image, convert the map image into a grayscale image, and use high-speed blur to smooth the map image and reduce noise; Use the Canny edge detection algorithm to identify edges in map images; use the OpenCV library to detect contours in map images; Perform feature extraction on the detected contour to obtain the area, perimeter, width, length and circularity of the contour; Set a contour area screening threshold to filter contours with an area smaller than the screening threshold; perform curve contour feature extraction on the remaining contours to obtain curve contour position features, length features, and curvature features; The extracted curve contour is assigned an identification label; the width and length characteristics of the tunnel are determined using the identified contour width and length, and the angle characteristics of the tunnel curve are determined using the identified contour curvature characteristics.

[0021] 2. Generate the initial inspection route plan based on the inspection task data, including the following steps: Acquire path generation information, wherein the path generation information includes tunnel entrance location information, tunnel exit location information, tunnel length information, curve location information, and curve angle information; Using the A-star algorithm, a preliminary inspection path from the tunnel entrance and tunnel exit is generated based on the acquired path generation information.

[0022] 3. Collecting environmental data outside the tunnel robot and the motion state data of the tunnel robot, wherein the environmental data includes one or more of distance data, thermal imaging data, video data, and audio data. Specifically, obtaining point cloud data of the external environment of the tunnel robot through a multi-line laser radar; The infrared thermal imaging device is used to obtain thermal images of the tunnel robot's external environment. The high-definition camera is used to obtain high-definition images of the tunnel robot's external environment. The ultrasonic ranging sensor is used to obtain the distance data between the tunnel robot and objects inside the tunnel. The gas monitoring sensor is used to obtain the gas concentration data outside the tunnel robot. The inertial measurement unit and speed sensor are used to obtain the motion state data of the tunnel robot.

[0023] 4. Use the external environmental data of the robot to model the tunnel environment, combine it with the motion state data, and autonomously generate state adjustment instructions to adjust the operating parameters of the tunnel robot.

[0024] The process of tunnel environment modeling is as follows: Read point cloud data, remove point cloud data noise, and then perform downsampling and filtering.

[0025] Obtain tunnel distribution characteristics; obtain infrared thermal images, and perform denoising and contrast enhancement on the infrared thermal images.

[0026] The point cloud data and infrared imaging data are registered to align the point cloud data and infrared imaging data in space; the point cloud data is used to generate a three-dimensional network model inside the tunnel, and the infrared thermal imaging data is used to add texture information to the three-dimensional model.

[0027] The Open3D library is used to visualize the three-dimensional network model inside the tunnel.

[0028] The three-dimensional network model inside the tunnel is generated by using the point cloud data, including the following steps: Using a normal vector estimation algorithm, key features inside the tunnel are identified, including tunnel width, tunnel floor, and tunnel ceiling.

[0029] The point cloud data is segmented by clustering algorithm or region growing algorithm, and the segmented point cloud data is classified by classifier to distinguish different key features.

[0030] According to the classification results, the point cloud data inside the tunnel is reconstructed in 3D using the triangulation method to generate a 3D network model inside the tunnel.

[0031] Then, based on the three-dimensional network model containing obstacle information, the initial inspection path planning is updated as follows: Import a 3D network model and divide it into multiple 3D grids.

[0032] Set up a 3D array for each 3D grid ,in, , Represents the total number of three-dimensional grids; They respectively represent the coordinates of the three-dimensional grid in the spatial coordinate system.

[0033] Traverse each 3D grid and check whether each 3D grid intersects with the obstacle. If it intersects with the obstacle, ,otherwise .

[0034] Based on the assignment of the three-dimensional array, the A-star algorithm is used to find the path from the starting point to the end point of the three-dimensional network model to update the initial inspection path. The updated inspection path is then sent to the tunnel robot.

[0035] The process of adjusting the operating parameters is as follows: The current position information of the robot is obtained; the current motion state data of the robot is obtained, and the motion state data includes the motion speed, acceleration, angular velocity and posture of the robot.

[0036] Put the robot in The position information, motion speed, acceleration, angular velocity and posture data at the moment constitute the motion state feature set ,in Respectively The position vector, velocity vector, acceleration vector, angular velocity vector and attitude vector of the moment tunnel robot; , , , , ;in, represents the spatial coordinates of the tunnel robot, Tunnel robot Always Axis speed, The speed of the axis and Speed ​​in the axis direction; Tunnel robot Always The acceleration in the axial direction, The acceleration in the axial direction and Acceleration in the axis direction; Tunnel robot Always Angular velocity in the direction of the axis, The angular velocity of the axis and Angular velocity in the direction of the axis; Tunnel robot Roll angle, pitch angle and yaw angle at the moment; based on , through Kalman filter, predicting tunnel robot The state of motion at all times; Implemented through particle filter Data fusion in. Particle filter is applied to multi-sensor data fusion, which can effectively deal with the uncertainty and noise between different sensors and improve the accuracy of state estimation.

[0037] According to the predicted motion state, the tunnel robot control instructions are generated to control the tunnel robot to accelerate, turn, etc., so that it can inspect along the updated inspection path.

[0038] In addition, in this embodiment, the data measured by the gas concentration sensor and the infrared thermal imaging data can be fused. Specifically, multiple features are extracted from the two types of data: maximum gas concentration, temperature distribution and hot spot area features, and the data are fused using the weighted average method to form a fused feature data set; After normalizing the fused feature data set, the pre-trained random forest classification model is used to predict environmental parameters, thereby helping staff to obtain the development trend of temperature and gas concentration inside the tunnel and determine whether appropriate measures need to be taken to adjust the environmental parameters in the tunnel to ensure the cable life and the working efficiency of the robot.

[0039] The present invention also provides a multi-sensor data fusion navigation system for a cable tunnel robot, comprising a multi-sensor data acquisition module, a data fusion processing module, an autonomous navigation control module, a motion module and a feedback control module, wherein the multi-sensor data acquisition module is used to acquire environmental data around the tunnel robot, the data fusion processing module processes the environmental data, the autonomous navigation control module makes path planning and motion instructions according to the processing results, the motion module is used to control the motion state of the tunnel robot according to the path planning and motion instructions, and the feedback control module is used to provide feedback on the motion instructions of the tunnel robot.

[0040] The multi-sensor data acquisition module includes a laser radar, an infrared thermal imaging device, a high-definition camera, an inertial measurement unit (IMU), a speed sensor, an ultrasonic sensor and a gas detector.

[0041] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the multi-sensor data fusion navigation method of a cable tunnel robot.

[0042] In the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. The embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present application are executed. It should be noted that the computer-readable medium mentioned above in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wire segments, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, electrical wire, optical cable, RF, etc., or any suitable combination of the foregoing.

[0043] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in a different order than the order marked in the accompanying drawings. For example, two boxes represented in succession 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 flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0044] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the principles, the implementation methods of the present invention may be subject to any changes or modifications.

Claims

1. A cable tunnel robot multi-sensor data fusion navigation method, characterized in that: The method comprises: Obtain inspection task data of the tunnel robot, and extract tunnel distribution characteristics, tunnel entrance location, tunnel exit location, task starting point location information, inspection task point location information, and task end point location information from the inspection task data; Generate initial inspection route planning based on inspection task data; Collecting environmental data outside the tunnel robot and motion state data of the tunnel robot; the environmental data includes one or more of distance data, thermal imaging data, video data, and audio data; The tunnel environment is modeled using the external environmental data of the tunnel robot. Combined with the motion state data, state adjustment instructions are generated autonomously to adjust the operating parameters of the tunnel robot.

2. A cable tunnel robot multi-sensor data fusion navigation method according to claim 1, characterized in that: The step of extracting tunnel distribution features from inspection task data comprises the following steps: Obtain the tunnel distribution map image, convert the map image into a grayscale image, and use high-speed blur to smooth the map image and reduce noise; Identify edges in map images using the Canny edge detection algorithm; Detect contours in map images using the OpenCV library; Perform feature extraction on the detected contour to obtain the area, perimeter, width, length and circularity of the contour; Set the contour area screening threshold to filter contours whose area is smaller than the screening threshold; Perform curve contour feature extraction on the remaining contours to obtain curve contour position features, length features and curvature features; Assigning identification labels to the extracted curve profiles; The width and length of the tunnel are determined using the identified contour width and length, and the angle characteristics of the tunnel curve are determined using the identified contour curvature characteristics.

3. A cable tunnel robot multi-sensor data fusion navigation method according to claim 2, characterized in that: Generating the initial inspection path planning according to the inspection task data includes the following steps: Acquire path generation information, wherein the path generation information includes tunnel entrance location information, tunnel exit location information, tunnel length information, curve location information, and curve angle information; The path search algorithm A-star is used to generate preliminary inspection paths from the tunnel entrance and tunnel exit based on the acquired path generation information.

4. A cable tunnel robot multi-sensor data fusion navigation method according to claim 3, characterized in that: The collecting of the external environment data of the tunnel robot and the motion state data of the tunnel robot comprises the following steps: Acquire point cloud data of the external environment of the tunnel robot through multi-line laser radar; Acquire thermal images of the external environment of the tunnel robot through an infrared thermal imaging device; Obtain high-definition images of the external environment of the tunnel robot through a high-definition camera; The distance data between the tunnel robot and the objects inside the tunnel is obtained through the ultrasonic ranging sensor; Obtain gas concentration data outside the tunnel robot through gas monitoring sensors; The motion state data of the tunnel robot is obtained through the inertial measurement unit and speed sensor.

5. A cable tunnel robot multi-sensor data fusion navigation method according to claim 4, characterized in that: The method of using the external environment data of the tunnel robot to model the tunnel environment includes the following steps: Read point cloud data, remove point cloud data noise, and then perform downsampling and filtering; Obtain tunnel distribution characteristics; Acquire infrared thermal images, perform denoising and contrast enhancement on the infrared thermal images; The point cloud data and the infrared imaging data are registered to align the point cloud data and the infrared imaging data in space; Use point cloud data to generate a 3D network model of the tunnel interior, and use infrared thermal imaging data to add texture information to the 3D model; Use Open3D library to visualize the 3D network model inside the tunnel; The method of generating a three-dimensional network model inside the tunnel using point cloud data comprises the following steps: Using a normal vector estimation algorithm, identifying key features inside the tunnel, including tunnel width, tunnel floor, and tunnel ceiling; The point cloud data is segmented by clustering algorithm or region growing algorithm, and the segmented point cloud data is classified by classifier to distinguish different key features; According to the classification results, the point cloud data inside the tunnel is reconstructed in 3D using the triangulation method to generate a 3D network model inside the tunnel.

6. A cable tunnel robot multi-sensor data fusion navigation method according to claim 5, characterized in that: The tunnel environment modeling using the environment data outside the tunnel robot also includes the following steps: Obstacle point cloud data is obtained through multi-line laser radar; Use the density clustering DBSCAN algorithm to perform cluster analysis on obstacle point cloud data and identify different obstacles; Perform feature extraction on the obstacle point cloud data after cluster analysis to obtain the position, size, shape and volume features of different obstacles to form feature data sets of different obstacles; Importing the three-dimensional network model inside the tunnel, obtaining multiple node coordinates of the three-dimensional network model, traversing each node of the three-dimensional network model, and checking whether each node intersects with the obstacle to determine whether the obstacle intersects with the tunnel or is located inside the tunnel; According to the intersection position between the node and the obstacle, the obstacle information is added to the three-dimensional network model inside the tunnel, and the node status is updated, wherein the node status includes marking as requiring detour or impassable; Output the three-dimensional network model containing obstacle information.

7. A cable tunnel robot multi-sensor data fusion navigation method according to claim 6, characterized in that: Based on the three-dimensional network model containing obstacle information, the initial inspection path planning is updated, including the following steps: Import a 3D network model and divide the 3D network model into multiple 3D grids; Set up a 3D array for each 3D grid ,in, , Represents the total number of three-dimensional grids; They represent the coordinates of the three-dimensional grid in the spatial coordinate system respectively; Traverse each 3D grid and check whether each 3D grid intersects with the obstacle. If it intersects with the obstacle, ,otherwise ; Based on the assignment of the three-dimensional array, the path search algorithm A-star is used to find the path from the starting point to the end point of the three-dimensional network model to update the initial inspection path; And send the updated inspection path to the tunnel robot.

8. A cable tunnel robot multi-sensor data fusion navigation method according to claim 7, characterized in that: The method of combining the motion state data, autonomously generating a state adjustment instruction, and adjusting the operating parameters of the tunnel robot includes the following steps: Get the current location information of the tunnel robot; Acquire current motion state data of the tunnel robot, wherein the motion state data includes motion speed, acceleration, angular velocity and posture of the tunnel robot; The tunnel robot The position information, motion speed, acceleration, angular velocity and posture data at the moment constitute the motion state feature set ,in Respectively The position vector, velocity vector, acceleration vector, angular velocity vector and attitude vector of the moment tunnel robot; , , , , ;in, represents the spatial coordinates of the tunnel robot, Tunnel robot Always Axis speed, The speed of the axis and Speed ​​in the axis direction; Tunnel robot Always The acceleration in the axial direction, The acceleration in the axial direction and Acceleration in the axis direction; Tunnel robot Always Angular velocity in the direction of the axis, The angular velocity of the axis and Angular velocity in the direction of the axis; Tunnel robot Roll angle, pitch angle and yaw angle at the moment; based on , through Kalman filter, predicting tunnel robot The state of motion at all times; Implemented through particle filter Data fusion in According to the predicted motion state, a tunnel robot control instruction is generated to control the tunnel robot to accelerate and / or turn.

9. A cable tunnel robot multi-sensor data fusion navigation system, characterized in that: It includes a multi-sensor data acquisition module, a data fusion processing module, an autonomous navigation control module, a motion module and a feedback control module. The multi-sensor data acquisition module is used to obtain environmental data around the tunnel robot. The data fusion processing module performs fusion processing on the environmental data. The autonomous navigation control module makes path planning and motion instructions according to the fusion processing results. The motion module is used to control the motion state of the tunnel robot according to the path planning and motion instructions. The feedback control module is used to provide feedback on the motion instructions of the tunnel robot.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement a cable tunnel robot multi-sensor data fusion navigation method as described in any one of claims 1 to 8.

Images