Autonomous navigation method and system of foot-type robot for comprehensive pipeline corridor inspection

Abstract

The present invention discloses a method and system for autonomous navigation of a legged robot for comprehensive pipeline corridor inspection, comprising the following steps: obtaining three-dimensional point cloud information of the comprehensive pipeline corridor and preprocessing the three-dimensional point cloud information; segmenting the preprocessed three-dimensional point cloud information into N planes, and obtaining normal vectors for each of the N planes; using a particle filter algorithm based on the obtained normal vectors to detect the extension direction of the pipeline corridor in real time; and controlling the inspection robot to move along the extension direction of the pipeline corridor, thereby achieving autonomous navigation of the inspection robot without a global map. The present invention uses a particle filter algorithm to detect the extension direction of the pipeline corridor in real time, and the inspection robot's navigation no longer relies on GPS and pre-established maps, thus solving the problem of low absolute positioning accuracy caused by repetitive scenarios.

Description

Technical Field

[0001] The present invention relates to the field of robot inspection technology, and in particular to an autonomous navigation method and system for a legged robot used for comprehensive pipeline corridor inspection. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Urban underground utility corridors are public pipeline tunnels built beneath cities, centrally arranging two or more municipal pipelines for electricity, communications, gas, water, heat, and drainage, among other services, with unified planning, design, construction, and maintenance. These corridors can range from a few kilometers to tens of kilometers in length, with narrow interiors and relatively closed spaces. Manual inspections are inefficient and present numerous safety risks.

[0004] The emergence of integrated pipeline corridor inspection robots enables uninterrupted monitoring of the integrated pipeline corridor environment, equipment, security, access control, etc., as well as disaster warning and disposal. Due to the narrow and long interior of the pipeline corridor and the lack of GPS signals, the robot cannot complete automatic navigation by first building a map and then real-time positioning. In addition, the scenes in the pipeline corridor are repetitive, and whether using lidar or visual positioning, mismatching is prone to occur, making it difficult to obtain the robot's accurate position along the pipeline corridor. Some pipeline corridor inspection robots use the method of installing tracks to enable the robot to move along the track, and positioning is achieved by adding RFID tags or barcodes to the track. However, due to the narrow and long pipeline corridor, the cost of installing the track is much higher than the cost of the robot itself. Summary of the Invention

[0005] In order to solve the above problems, the present invention proposes an autonomous navigation method and system for a legged robot for integrated pipeline corridor inspection, which can overcome the problem of low absolute positioning accuracy caused by repetitive scenes without relying on GPS signals and realize autonomous navigation of the robot.

[0006] In some embodiments, the following technical solutions are adopted:

[0007] A method for autonomous navigation of a foot-type robot for comprehensive pipe gallery inspection, comprising:

[0008] Obtain the 3D point cloud information of the integrated pipeline corridor and pre-process the 3D point cloud information;

[0009] The pre-processed 3D point cloud information is divided into N planes, and the normal vectors of the N planes are obtained respectively;

[0010] Based on the obtained normal vector, the particle filter algorithm is used to detect the extension direction of the pipeline corridor in real time;

[0011] The inspection robot is controlled to move along the extension direction of the pipeline corridor to realize autonomous navigation of the inspection robot without a global map.

[0012] Among them, based on the obtained normal vector, the particle filter algorithm is used to detect the extension direction of the pipeline corridor in real time. The specific process is as follows:

[0013] Initialize M three-dimensional vectors and normalize the three-dimensional vectors;

[0014] Each normalized three-dimensional vector is regarded as a particle, and the weight of the particle is calculated;

[0015] The particle with the largest weight among all particles is selected. If the weight of the particle exceeds the set threshold, the direction corresponding to the particle is the extension direction of the corridor. If the weight of the particle does not exceed the set threshold, the extension direction of the corridor is ambiguous. The scanning distance of the integrated corridor 3D point cloud is reduced and the extension direction of the corridor is re-determined. If the maximum particle weight still does not exceed the set threshold, it is determined that the corridor has reached its end point.

[0016] The method for calculating particle weight is:

[0017]

[0018] Among them, n i Represents the normal vector obtained for each plane, i is an integer from 1 to N; k jn is the normalized three-dimensional vector, p i Indicates the number of point clouds in each plane.

[0019] The threshold value is set to q is the threshold coefficient, q∈(0,1).

[0020] In other embodiments, the following technical solutions are adopted:

[0021] A foot-type robot autonomous navigation system for comprehensive pipeline corridor inspection, comprising:

[0022] 3D point cloud acquisition module, used to obtain 3D point cloud information of the integrated pipeline corridor and pre-process the 3D point cloud information;

[0023] The point cloud segmentation module is used to segment the pre-processed 3D point cloud information into N planes and obtain the normal vectors for each of the N planes;

[0024] The pipeline corridor extension direction detection module is used to detect the extension direction of the pipeline corridor in real time based on the obtained normal vector and the particle filter algorithm;

[0025] The robot navigation module is used to control the inspection robot to move along the extension direction of the pipeline corridor, realizing the inspection robot's autonomous navigation without a global map.

[0026] In other embodiments, the following technical solutions are adopted:

[0027] A patrol robot is characterized in that the above-mentioned legged robot autonomous navigation method for integrated pipeline corridor inspection is used to navigate the patrol robot.

[0028] In other embodiments, the following technical solutions are adopted:

[0029] A terminal device includes a processor and a memory, the processor is used to implement various instructions; the memory is used to store multiple instructions, and the instructions are suitable for being loaded and executed by the processor to implement the above-mentioned legged robot autonomous navigation method for integrated pipeline corridor inspection.

[0030] In other embodiments, the following technical solutions are adopted:

[0031] A computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded and executed by a processor of a terminal device to implement the above-mentioned autonomous navigation method of a legged robot for integrated pipeline corridor inspection.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] (1) The present invention proposes a method for autonomous navigation of an integrated pipeline corridor inspection robot without a global map, develops a particle filter detection algorithm for the pipeline corridor extension direction, and realizes the autonomous navigation of the inspection robot without a global map. The inspection robot navigation no longer relies on GPS and pre-established maps, and solves the problem of low absolute positioning accuracy caused by repeated scenes. At the same time, the inspection robot is no longer limited to a track inspection robot, eliminating the cost of setting up inspection tracks, improving the adaptability of the inspection robot to integrated pipeline corridor inspection, and improving the positioning accuracy of the inspection robot.

[0034] Other features and advantages of additional aspects of the present invention will be given in part in the following description and in part will become obvious from the following description or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 The present invention provides a flowchart of the autonomous navigation method without a global map for the integrated pipeline corridor inspection robot in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0037] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0038] Example 1

[0039] In one or more embodiments, a method for autonomous navigation of a foot-type robot for comprehensive pipeline corridor inspection is disclosed, combining Figure 1 , specifically including the following process:

[0040] S101: Acquire the three-dimensional point cloud information of the integrated pipeline corridor and pre-process the three-dimensional point cloud information;

[0041] In this embodiment, a sensor is used to scan the environmental contour to obtain three-dimensional point cloud information of the integrated pipeline corridor within a set distance range; the sensor can be a binocular vision camera, a structured light camera, a TOF camera or a three-dimensional lidar, etc.

[0042] After obtaining the 3D point cloud information, filters are used to preprocess the 3D point cloud, including: using statistical filters or radius filters to remove isolated points, and using voxel filters to downsample to avoid excessive concentration of point clouds.

[0043] S102: Segment the pre-processed 3D point cloud information into N planes, and obtain normal vectors for each of the N planes;

[0044] In this embodiment, the RANSAC method is used to perform plane segmentation on the point cloud to obtain N planes (mainly walls in the pipe gallery, and the number of point clouds on other planes is relatively small). The number of points in each plane is recorded as p. i , where i is an integer from 1 to N. It should be noted that the process of plane segmentation of point cloud using RANSAC method is a technology known to those skilled in the art and will not be described in detail.

[0045] Then calculate the normal vectors for the above N planes respectively and get the normal vector n i , where i is an integer from 1 to N.

[0046] S103: Based on the obtained normal vector, a particle filter algorithm is used to detect the extension direction of the pipeline corridor in real time;

[0047] Since the extension direction of the pipe gallery should be perpendicular to the normal vectors of most walls, this embodiment uses a particle filter algorithm to calculate the extension direction of the pipe gallery. The specific process is as follows:

[0048] S1031: Initialize M three-dimensional vectors, expressed as where x j 、y j 、z j They are all initialized to random numbers, j is an integer from 1 to M.

[0049] S1032: Normalize the non-zero vectors. For the zero vector, reinitialize it with random numbers and then normalize it.

[0050] S1033: For each normalized k jn , as a particle; find the corresponding particle weight F j :

[0051]

[0052] Among them, n i ·k jn Represents the inner product of the particle vector and the wall normal vector (plane normal vector). The smaller the value, the more perpendicular the two vectors are. Indicates the matching difference between the particle and a single wall. The more points in the wall point cloud and the more perpendicular the particle vector is to the wall normal, the smaller the difference. The sum of the differences between each particle and each wall is taken as the reciprocal, which is the weight of the particle in the particle filter algorithm.

[0053] S1034: Select the particle with the largest weight among the particles. If the weight of the particle exceeds the set threshold, the corresponding The direction of the vector corresponding to the particle is the corridor extension direction. Here, q is the threshold coefficient, q∈(0,1); the closer q is to 0, the stricter the screening, while q closer to 1 indicates a looser screening. The specific value of q can be set according to actual needs. If the particle weight does not exceed the threshold, it is determined that the corridor extension direction is ambiguous. By reducing the distance of the sensor scanning the point cloud, the scanning range is limited to a closer range, and the above steps are repeated to determine the corridor extension direction in the vicinity. If it still does not exceed the threshold, it is determined that the corridor has reached its end point.

[0054] Because the tunnel's direction doesn't change dramatically, the next particle placement can be done near the previous cycle's results, enabling iterative search. Therefore, each particle is iteratively updated according to its weight. Through repeated iterations, a continuous and progressive output is achieved within each cycle, enabling real-time detection of the tunnel's extension direction.

[0055] S104: Control the inspection robot to move along the extension direction of the pipeline corridor to achieve autonomous navigation of the inspection robot without a global map.

[0056] The inspection robot navigation method of this embodiment no longer relies on GPS and pre-established maps, which solves the problem of low absolute positioning accuracy caused by repetitive scenarios. At the same time, the inspection robot is no longer limited to track inspection robots, eliminating the cost of setting up inspection tracks and expanding the applicable types of inspection robots. For example, legged robots can be used to inspect underground integrated pipelines, and the positioning accuracy of the inspection robot is also improved.

[0057] Example 2

[0058] In one or more embodiments, a legged robot autonomous navigation system for integrated pipeline corridor inspection is disclosed, comprising:

[0059] 3D point cloud acquisition module, used to obtain 3D point cloud information of the integrated pipeline corridor and pre-process the 3D point cloud information;

[0060] The point cloud segmentation module is used to segment the pre-processed 3D point cloud information into N planes and obtain the normal vectors for each of the N planes;

[0061] The pipeline corridor extension direction detection module is used to detect the extension direction of the pipeline corridor in real time based on the obtained normal vector and the particle filter algorithm;

[0062] The robot navigation module is used to control the inspection robot to move along the extension direction of the pipeline corridor, realizing the inspection robot's autonomous navigation without a global map.

[0063] It should be noted that the specific implementation of the above modules is the same as that in Example 1 and will not be repeated here.

[0064] Example 3

[0065] In one or more embodiments, a patrol robot is disclosed, which uses the autonomous navigation method of a legged robot for integrated pipeline corridor inspection described in Example 1 to navigate the patrol robot.

[0066] Example 4

[0067] In one or more embodiments, a terminal device is disclosed, including a server. The server includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the legged robot autonomous navigation method for integrated utility corridor inspection described in Example 1 is implemented. For the sake of brevity, this description is omitted here.

[0068] It should be understood that in this embodiment, the processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0069] The memory may include a read-only memory and a random access memory, and provides instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0070] During implementation, each step of the above method may be completed by an integrated logic circuit of hardware in a processor or by instructions in the form of software.

[0071] In other embodiments, a computer-readable storage medium is disclosed, in which a plurality of instructions are stored, wherein the instructions are suitable for being loaded and executed by a processor of a terminal device to implement the global map-free autonomous navigation method of the integrated pipeline corridor inspection robot in Example 1.

[0072] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. A method for autonomous navigation of a foot-type robot for comprehensive pipeline corridor inspection, characterized in that: include: Obtain the 3D point cloud information of the integrated pipeline corridor and pre-process the 3D point cloud information; The pre-processed 3D point cloud information is divided into N planes, and the normal vectors of the N planes are obtained respectively; Based on the obtained normal vector, the particle filter algorithm is used to detect the extension direction of the pipeline corridor in real time. The specific process is as follows: Initialize M three-dimensional vectors, expressed as ,in 、 、 are initialized to random numbers. is an integer from 1 to M; Normalize non-zero vectors, , for the zero vector, re-initialize it with random numbers and then normalize it; For each unitized , as a particle, find the corresponding particle weight : in, is an integer from 1 to N; Represents the normal vector obtained for each plane; Indicates the number of point clouds in each plane; Represents the inner product of the particle vector and the wall normal vector. The smaller the value, the more perpendicular the two vectors are. Indicates the matching difference between the particle and a single wall. The more points in the wall point cloud and the more perpendicular the particle vector is to the wall normal vector, the smaller the difference. The sum of the differences between each particle and each wall, after taking the reciprocal, is the weight of the particle in the particle filter algorithm; Take the particle with the largest weight among the particles. If the weight of the particle exceeds the set threshold, the corresponding , then the direction of the vector corresponding to the particle is the extension direction of the pipeline corridor, where is the threshold coefficient, ; The inspection robot is controlled to move along the extension direction of the pipeline corridor to realize autonomous navigation of the inspection robot without a global map.

2. The autonomous navigation method of a legged robot for comprehensive pipeline corridor inspection according to claim 1, characterized in that: Preprocess the 3D point cloud information, including: using statistical filters or radius filters to remove isolated points, and using voxel filters to downsample.

3. The autonomous navigation method of a legged robot for comprehensive pipeline corridor inspection according to claim 1, characterized in that: The RANSAC method is used to segment the preprocessed 3D point cloud information into N planes.

4. A foot-type robot autonomous navigation system for comprehensive pipeline corridor inspection, characterized in that: include: 3D point cloud acquisition module, used to obtain 3D point cloud information of the integrated pipeline corridor and pre-process the 3D point cloud information; The point cloud segmentation module is used to segment the pre-processed 3D point cloud information into N planes and obtain the normal vectors for each of the N planes; The pipeline corridor extension direction detection module is used to detect the extension direction of the pipeline corridor in real time based on the obtained normal vector and the particle filter algorithm. The specific process is as follows: Initialize M three-dimensional vectors, expressed as ,in 、 、 are initialized to random numbers. is an integer from 1 to M; Normalize non-zero vectors, , for the zero vector, re-initialize it with random numbers and then normalize it; For each unitized , as a particle, find the corresponding particle weight : in, is an integer from 1 to N; Represents the normal vector obtained for each plane; Indicates the number of point clouds in each plane; Represents the inner product of the particle vector and the wall normal vector. The smaller the value, the more perpendicular the two vectors are. Indicates the matching difference between the particle and a single wall. The more points in the wall point cloud and the more perpendicular the particle vector is to the wall normal vector, the smaller the difference. The sum of the differences between each particle and each wall, after taking the reciprocal, is the weight of the particle in the particle filter algorithm; Select the particle with the largest weight among the particles. If the weight of the particle exceeds the set threshold, the corresponding , then the direction of the vector corresponding to the particle is the extension direction of the pipeline corridor, where is the threshold coefficient, ; The robot navigation module is used to control the inspection robot to move along the extension direction of the pipeline corridor, realizing the inspection robot's autonomous navigation without a global map.

5. A patrol robot, characterized in that: The inspection robot is navigated by adopting the autonomous navigation method of a legged robot for integrated pipeline corridor inspection as described in any one of claims 1 to 3.

6. A terminal device comprising a processor and a memory, wherein the processor is used to implement various instructions; the memory is used to store multiple instructions, characterized in that: The instructions are suitable for being loaded by a processor and executing the autonomous navigation method of a legged robot for integrated pipeline corridor inspection as described in any one of claims 1-3.

7. A computer-readable storage medium storing a plurality of instructions, characterized in that: The instructions are suitable for being loaded by the processor of the terminal device and executing the autonomous navigation method of the legged robot for integrated pipeline corridor inspection as described in any one of claims 1-3.

Images