Subway tunnel inspection robot comprehensive positioning method and system

Abstract

The application discloses a kind of subway tunnel inspection robot integrated positioning method and system, servo motor encoder is installed on inspection robot, inertial sensor, camera, RFID label, RFID receiver is installed on both sides of each platform of subway tunnel, multiple adjacent setting distance line marks are evenly provided on the line between every two platforms of subway tunnel, and the method is as follows: real-time acquisition tire rotation number of inspection robot, robot motion parameter, the tunnel image of shooting and the RFID platform signal fed back when inspection robot passes through platform;Further, line mark information and the pose information of robot are calculated and acquired;According to RFID platform signal, tire rotation number, line mark information and pose information, the integrated positioning of inspection robot is carried out, and position and pose information are output.The present application realizes efficient, accurate, implementable centimeter-level error positioning to inspection robot by establishing fusion identification rule and integrated positioning method.

Description

Technical Field

[0001] This invention relates to the field of robot positioning technology, and in particular to a comprehensive positioning method and system for a subway tunnel inspection robot. Background Technology

[0002] With the advancement of technology and the development of the times, subways have emerged as a convenient and efficient mode of transportation. Due to their advantages such as saving land, reducing surface noise and pollution, high speed and punctuality, and large capacity, subway transportation has broad development prospects and a continuously expanding market. However, as the mileage of subway construction continues to increase, its operation and maintenance have become a prominent issue. Traditional manual inspection and maintenance methods are not only inefficient and time-consuming but also highly subjective and lack sensitivity. With the development of computers, subway tunnel inspection robots equipped with intelligent algorithms are gradually replacing manual tunnel inspections.

[0003] However, because subway tunnels are constructed of reinforced concrete and other materials, these materials shield satellite signals, preventing GNSS (Global Navigation Satellite System) signals from penetrating the tunnel walls. Even if GNSS signals can penetrate the tunnel walls, the multipath effect within the tunnel (i.e., the signal is reflected and refracted by walls, the ground, and other objects during propagation) causes severe signal attenuation, making it impossible to receive and process them properly, and thus impossible to locate the subway tunnel inspection robot. Therefore, how to locate the subway tunnel inspection robot (hereinafter referred to as "inspection robot") in the absence of GNSS signals has become a key issue in the development of intelligent tunnels.

[0004] When existing inspection robots perform tasks, the system typically needs to follow a pre-set map as the inspection path to reach the task point. During movement, the system acquires data from the odometer installed on the robot's wheels for positioning information. However, because odometer technology is affected by tire wear, track slippage, and encoder accuracy, this technology is more suitable for short-range positioning and cannot achieve long-distance robot positioning in subway tunnels. In addition, there are currently positioning methods based on LiDAR, but this method is easily affected by environmental factors, resulting in poor positioning accuracy. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, this invention provides a comprehensive positioning method and system for subway tunnel inspection robots. By establishing fusion identification rules and a comprehensive positioning method, the invention achieves the effective integration of multiple positioning methods. The comprehensive positioning method is used to locate the inspection robot, enabling efficient, accurate, and feasible centimeter-level error positioning of the inspection robot. This solves the problem that traditional single positioning methods cannot achieve efficient and accurate positioning of equipment in subway tunnels.

[0006] In a first aspect, the present invention provides a comprehensive positioning method for a subway tunnel inspection robot.

[0007] A comprehensive positioning method for a subway tunnel inspection robot, comprising: the inspection robot equipped with a servo motor encoder, inertial sensor, camera, and RFID tag; RFID receivers installed on both sides of each platform in the subway tunnel; and multiple adjacent line markers evenly spaced at predetermined intervals along the track between every two platforms in the subway tunnel. The method includes:

[0008] The system can acquire real-time data on the number of tire rotations of the inspection robot, robot motion parameters, captured tunnel images, and RFID platform signals fed back by the inspection robot as it passes through the platform.

[0009] Based on the captured tunnel images, a target detection algorithm is used to identify the route markers that appear during the robot's inspection process and obtain the route marker information; based on the robot's motion parameters, integral calculations are performed to calculate the robot's pose information.

[0010] Based on real-time acquired RFID platform signals, tire rotation count, route marking information, and pose information, the inspection robot is comprehensively positioned, and its real-time location and pose information are output.

[0011] Secondly, the present invention provides a comprehensive positioning system for a subway tunnel inspection robot.

[0012] A comprehensive positioning system for a subway tunnel inspection robot includes an inspection robot and a back-end positioning system. The inspection robot is equipped with a servo motor encoder, an inertial sensor, a camera, and an RFID tag. RFID receivers are installed on both sides of each platform in the subway tunnel. Multiple adjacent line markers with a set distance are evenly distributed on the line between every two platforms in the subway tunnel.

[0013] The servo motor encoder, inertial sensor, camera, RFID tag and RFID receiver are used to acquire in real time the number of tire rotations of the inspection robot, robot motion parameters, captured tunnel images, and RFID platform signals fed back by the inspection robot when it passes the platform.

[0014] The inspection robot is equipped with an onboard computer, which is connected to the background positioning system to receive acquired data information, perform preliminary processing on the data information, and transmit the preliminarily processed data information to the background positioning system.

[0015] The background positioning system is used to perform comprehensive positioning of the inspection robot based on real-time acquired RFID platform signals, tire rotation count, route marking information, and pose information, and output the real-time location and pose information of the inspection robot.

[0016] The above one or more technical solutions have the following beneficial effects:

[0017] 1. This invention provides a comprehensive positioning method and system for a subway tunnel inspection robot. By establishing fusion identification rules and a comprehensive positioning method, it achieves the effective integration of multiple positioning methods. It utilizes a comprehensive positioning method based on servo motor encoder and inertial navigation, a positioning method based on computer vision recognition of tunnel odometers, and an RFID-based interval positioning method to locate the inspection robot. This achieves efficient, accurate, and feasible centimeter-level error positioning of the inspection robot. It does not require complex equipment and is unaffected by the track road. The positioning is accurate, efficient, and rapid, solving the problem that traditional single positioning methods cannot achieve efficient and accurate positioning of equipment in subway tunnels.

[0018] 2. The integrated positioning method proposed in this invention uses an inertial sensor to acquire and process the pose information of the inspection robot. This inertial sensor does not rely on any external information and does not radiate energy to the outside, thus it is concealed and is not affected by external electromagnetic interference. This inertial sensor can quickly acquire pose information and update it in real time, so it is highly applicable to the real-time transmission of the pose of the inspection robot.

[0019] 3. The comprehensive positioning method proposed in this invention uses the identification of 100-meter markers along the tunnel and RFID positioning of the operating area to obtain absolute mileage information, and uses a servo motor encoder to obtain the number of wheel rotations to obtain relative mileage information. This combination of absolute and relative methods reduces positioning errors. The positioning equipment is simple, low-cost, and easy to apply, making it more convenient and efficient for rail transit maintenance operations. Furthermore, the method records the inspection robot's posture on the subway track in real time and uploads it to the cloud platform via a 5G network, facilitating comprehensive monitoring and control of the inspection process by technicians. It also records tunnel defects or anomalies detected by the inspection robot through other modules, marks their locations, and uploads photographic data for subsequent verification and repair by technicians. Attached Figure Description

[0020] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0021] Figure 1 This is a flowchart of the integrated positioning method for a subway tunnel inspection robot according to an embodiment of the present invention;

[0022] Figure 2 This is a schematic diagram of the integrated positioning method for subway tunnel inspection robots according to an embodiment of the present invention;

[0023] Figure 3 This is a schematic diagram illustrating the principle of the RFID positioning method in an embodiment of the present invention;

[0024] Figure 4 This is a schematic diagram illustrating an example of the integrated positioning method for a subway tunnel inspection robot in an embodiment of the present invention. Detailed Implementation

[0025] It should be noted that the following detailed descriptions are exemplary and are intended only to describe specific embodiments and to provide further explanation of the invention, and are not intended to limit the scope of exemplary embodiments of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0026] Example 1

[0027] This embodiment provides a comprehensive positioning method for a subway tunnel inspection robot. The positioning device is mounted on the inspection robot, and a comprehensive positioning method is adopted, which combines a positioning method based on servo motor encoder and inertial navigation, a positioning method based on computer vision recognition tunnel odometer, and a section positioning method based on RFID (Radio Frequency Identification) technology. This method can locate the inspection robot without the need for complex equipment, is not affected by the track road, and is accurate, efficient and rapid.

[0028] Specifically, the inspection robot is equipped with a positioning device, which includes a servo motor encoder, inertial sensor, camera, and RFID tag. RFID receivers are installed on both sides of each platform in the subway tunnel. Multiple adjacent line markers at predetermined intervals are evenly distributed along the track between every two platforms in the subway tunnel. The comprehensive positioning method used is as follows: Figure 1 As shown, it includes:

[0029] The system acquires real-time motion information of the inspection robot, captured tunnel images, and RFID platform signals fed back by the inspection robot as it passes the platform; the motion information includes the number of tire rotations and robot motion parameters.

[0030] Based on the captured tunnel images, target detection algorithms are used to identify line markers that appear during the robot's inspection process and obtain line marker information;

[0031] Based on the robot's motion parameters, integral calculations are performed to obtain the robot's pose information;

[0032] Based on real-time acquired RFID platform signals, tire rotation count, route marking information, and pose information, the inspection robot is comprehensively positioned, and its real-time location and pose information are output.

[0033] The following content provides a more detailed description of the integrated positioning method for subway tunnel inspection robots proposed in this embodiment.

[0034] Step S1: Pre-install positioning equipment. Install a positioning device on the inspection robot. This device includes a servo motor encoder, an inertial sensor, a camera, and an RFID tag. The servo motor encoder is mounted on the inspection robot and connected to its servo motor; the inertial sensor is mounted on the robot's chassis; the camera is mounted on the robot with its field of view facing the track markings; and the RFID tag is affixed to the surface of the tunnel inspection robot. RFID receivers are installed on both sides of each platform in the subway tunnel, and multiple adjacent track markings at predetermined intervals are evenly distributed along the track between every two platforms in the subway tunnel.

[0035] In this embodiment, the line marker is a rail transit line sign, which is installed on the right side of the line and is usually placed on the tunnel wall; the aforementioned set distance is 100m; furthermore, to ensure the safe inspection of the inspection robot, the distance between the camera and the 100m line marker is usually within 5-10m. In addition, the installation of the RFID receiver should not affect the normal operation of the subway.

[0036] Step S2: Obtain real-time positioning data. When the inspection robot is working, the servo motor encoder acquires and records the number of rotations of the robot's tires, and the inertial sensor acquires and records the robot's motion parameters, including speed, acceleration, and angular velocity. The camera captures tunnel images including track markings, and the RFID receiver detects the RFID receiver attached to the robot when it passes through a station, thereby obtaining the inspection robot's location information and providing feedback.

[0037] Specifically, inertial sensors include gyroscopes and accelerometers. These sensors provide feedback on robot motion parameters, including velocity, acceleration, and angular velocity. By integrating the acceleration information from the accelerometers and the rotational angular velocity information from the gyroscopes, the robot's pose changes can be calculated, thus obtaining its pose information for navigation and localization. Furthermore, the number of inertial sensors can be increased or decreased based on actual needs.

[0038] Furthermore, the servo motor encoder reads the number of tire rotations; the camera captures tunnel images and uses a target detection algorithm to identify line markers that appear during the robot's inspection process, obtaining line marker information; when the inspection robot passes by the RFID receivers installed on both sides of the station, the RFID tag attached to the inspection robot is detected by the receiver, which feeds back the RFID platform signal, that is, the location information of the inspection robot.

[0039] In this embodiment, as Figure 2 As shown, the inspection robot is equipped with an onboard computer, which is connected to a servo motor encoder, inertial sensors, and a camera. This computer receives and processes data such as the number of rotations of the robot's tires, robot motion parameters, and tunnel images, obtaining the robot's pose information and route markings. The onboard computer also communicates with a backend positioning system (a computer in this embodiment), transmitting the pre-processed data to the backend positioning system. Furthermore, as... Figure 3 As shown, the RFID receiver is connected to the background positioning system, and the RFID receiver can feed back the detected location information to the background positioning system.

[0040] Step S3: Upload the acquired real-time positioning data to the background positioning system. The background positioning system processes and calculates the data to comprehensively identify the real-time position and pose information of the inspection robot. In this embodiment, based on the real-time acquired RFID platform signals, tire rotation count, route marker information, and pose information, the inspection robot is comprehensively positioned, and the real-time position and pose information of the inspection robot are output, including:

[0041] Record the number of tire rotations from the moment the route sign information was last obtained to the current moment as C1; and if an RFID station signal is obtained during the recording of this parameter C1, clear the currently recorded number of tire rotations C1 and record it again.

[0042] Based on the line marker information, the number of line markers from the moment the RFID station signal was last acquired to the current moment is recorded as B;

[0043] Based on the latest RFID station signal acquired at the current moment, determine the station closest to the inspection robot, and combine the number of tire rotations C1 and the number of line markers B to locate the current position of the inspection robot.

[0044] Specifically, the real-time tire rotation count is saved and recorded as parameter C, which is continuously recorded. At the same time, the tire rotation count is also saved and recorded in parameter C1. When the inspection robot receives an RFID station signal or a route marker signal identified by the camera, parameter C1 is cleared and the count is restarted. That is, the tire rotation count from the moment the route marker information was last obtained to the current moment is recorded as C1. If an RFID station signal is obtained during the recording of parameter C1, the currently recorded tire rotation count is cleared and the count is restarted.

[0045] Based on the route marker information, the number of route markers from the moment the RFID station signal was last acquired to the current moment is recorded as B. That is, the route marker information is acquired using a computer vision-based positioning method, and this route marker information is the number of route markers B, which in this embodiment is the number of markers per 100 meters. This number of route markers B is reset to zero and re-recorded when the RFID station signal is acquired.

[0046] In addition, when an RFID signal is received, an RFID station signal is fed back, which is the serial number corresponding to the station that received the RFID signal; when no RFID signal is received, no feedback is given and the signal remains unchanged, and the fed-back RFID station signal is saved and recorded.

[0047] Based on the data signals saved and recorded above, the current position of the inspection robot is: distance from the Xth station is B*D+C1*L; where X represents the number of stations corresponding to the latest RFID station signal acquired at the current moment, B represents the number of line markers from the latest RFID station signal acquired to the current moment, D represents the set distance between adjacent line markers, in this embodiment, D=100, C1 represents the number of tire rotations from the moment the line marker information was acquired to the current moment, and L represents the wheel circumference of the inspection robot.

[0048] At the same time, the number of tire rotations from the initial inspection moment to the current moment is saved and recorded as C. Combined with the wheel circumference L of the inspection robot, the total distance traveled by the inspection robot is calculated as C*L.

[0049] As another implementation, this embodiment also provides a positioning method conflict correction method, in which positioning methods are hierarchically structured, and the priority of multiple positioning methods is defined. Specifically, the positioning methods used in this embodiment include a positioning method based on servo motor encoder and inertial navigation, a positioning method based on computer vision to identify tunnel mileage, and an RFID-based interval positioning method. The defined priority is: RFID-based interval positioning method > computer vision-based tunnel mileage identification positioning method > servo motor encoder and inertial navigation positioning method. While hierarchically structured the positioning methods, the lower-level hierarchical positioning method is given priority over the uppermost hierarchical positioning method.

[0050] In this embodiment, the principle of the above-mentioned conflict correction is as follows: if the current interval movement mileage exceeds the set distance between two line markers, it is considered that the acquisition of line marker information has failed, the line marker information is automatically supplemented, the current interval movement mileage is cleared and re-recorded; wherein, the current interval movement mileage is the product of the number of tire rotations C1 from the time when the latest line marker information was acquired to the current time and the wheel circumference L of the inspection robot, i.e., C1*L.

[0051] As another implementation, it also includes: if the encoder fails to record the number of tire rotations C1 from the moment the last line marker information was acquired to the current moment (e.g., encoder failure, stopping the output distance), the encoder is restarted after the next line marker information is received to re-record.

[0052] For example, such as Figure 4 As shown, the inspection robot begins its inspection task at platform A, acquiring RFID platform information (i.e., recording platform A). Simultaneously, based on a servo motor method, it continuously records the number of tire rotations C and C1. The number of tire rotations C and C1 are recorded in parallel. C is the number of tire rotations continuously recorded from the initial inspection time to the current time, which does not change with the input of other parameters. C1 is the number of tire rotations recorded from the time when the latest route marker information was acquired to the current time. The route marker information at the initial inspection time is 0, and the initial time when the latest route marker information was acquired is the initial inspection time. Therefore, the location result output for damage 1 is: distance from platform A: C1*L;

[0053] When the inspection robot passes through platform A, it identifies the first 100-meter marker and increments the parameter B count by one. C1 is reset to zero and the count is restarted and recorded again. C remains unchanged. At this time, the output location result of damage 2 is: distance from platform A: B*100+C1*L.

[0054] After the inspection robot passes the next station B, the RFID receiver identifies the RFID electronic tag attached to the inspection robot, feeds back the RFID station information, records station B, resets the 100-meter marker B to zero and starts counting again, resets C1 to zero and starts counting again, and keeps C unchanged. At this time, the output location result of damage 3 is: distance from station B: C1*L.

[0055] Step S4: During the inspection process, the inspection robot feeds back positioning data information according to a set time interval to perform comprehensive positioning of the inspection robot, thereby achieving positioning of the inspection robot throughout the entire inspection process. In this embodiment, the set time interval is 1 second.

[0056] As another implementation method, during the inspection process, based on the captured tunnel images, target detection algorithms are used to identify tunnel defects. If a tunnel defect is detected and identified, location data is immediately fed back to pinpoint its location for subsequent repair. Furthermore, while feeding back the location data, images of the tunnel defect are also fed back to assist in repair analysis.

[0057] This embodiment provides a comprehensive positioning method for a subway tunnel inspection robot. It uses an inertial sensor to acquire and process the robot's pose information. This inertial sensor does not rely on any external information and does not radiate energy to the outside, thus possessing concealment and being unaffected by external electromagnetic interference. This inertial sensor can quickly acquire pose information and update it in real time, making it very suitable for real-time transmission of the inspection robot's pose.

[0058] This embodiment provides a comprehensive positioning method for a subway tunnel inspection robot. It uses 100-meter markers along the tunnel and RFID positioning of the operating area to obtain absolute mileage information, and a servo motor encoder to obtain the number of wheel rotations to obtain relative mileage information. Combining absolute and relative methods reduces positioning errors. The positioning equipment is simple, low-cost, and easy to apply, making it more convenient and efficient for rail transit maintenance operations. Furthermore, the method described in this embodiment records the inspection robot's posture on the subway track in real time and uploads it to a cloud platform via a 5G network, facilitating comprehensive monitoring and control of the inspection process by technicians. It also records tunnel defects or anomalies detected by the inspection robot through other modules, marking their locations and uploading photographic data for subsequent verification and repair by technicians.

[0059] Example 2

[0060] A comprehensive positioning system for a subway tunnel inspection robot includes an inspection robot and a back-end positioning system. The inspection robot is equipped with a servo motor encoder, an inertial sensor, a camera, and an RFID tag. RFID receivers are installed on both sides of each platform in the subway tunnel. Multiple adjacent line markers with a set distance are evenly distributed on the line between every two platforms in the subway tunnel.

[0061] The servo motor encoder, inertial sensor, camera, RFID tag and RFID receiver are used to acquire in real time the number of tire rotations of the inspection robot, robot motion parameters, captured tunnel images, and RFID platform signals fed back by the inspection robot when it passes the platform.

[0062] The inspection robot is equipped with an onboard computer, which is connected to the background positioning system to receive acquired data information, perform preliminary processing on the data information, and transmit the preliminarily processed data information to the background positioning system.

[0063] The background positioning system is used to perform comprehensive positioning of the inspection robot based on real-time acquired RFID platform signals, tire rotation count, route marking information, and pose information, and output the real-time location and pose information of the inspection robot.

[0064] The steps and methods involved in the above embodiment two correspond to those in embodiment one. For specific implementation details, please refer to the relevant description section of embodiment one.

[0065] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0066] The above description is only a preferred embodiment of the present invention. Although the specific implementation of the present invention has been described 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, based on the technical solution of the present invention, various modifications or variations that can be made by those skilled in the art without creative effort are still within the scope of protection of the present invention.

Claims

1. A comprehensive positioning method for a subway tunnel inspection robot, characterized in that, The inspection robot is equipped with a servo motor encoder, inertial sensor, camera, and RFID tag. RFID receivers are installed on both sides of each platform in the subway tunnel. Multiple adjacent line markers are evenly distributed at predetermined intervals along the track between every two platforms in the subway tunnel. The method includes: The system can acquire real-time data on the number of tire rotations of the inspection robot, robot motion parameters, captured tunnel images, and RFID platform signals fed back by the inspection robot as it passes through the platform. Based on the captured tunnel images, a target detection algorithm is used to identify the route markers that appear during the robot's inspection process and obtain the route marker information; based on the robot's motion parameters, integral calculations are performed to calculate the robot's pose information. Based on the real-time acquired RFID platform signals, tire rotation count, line marking information, and pose information, the inspection robot is comprehensively positioned, and its real-time position and pose information are output. The priority of the positioning methods based on servo motor encoders and inertial navigation, the positioning methods based on computer vision to identify tunnel mileage, and the RFID-based interval positioning methods is as follows: RFID-based interval positioning method > Computer vision to identify tunnel mileage positioning method > Servo motor encoder and inertial navigation positioning method. Comprehensive positioning of the inspection robot in real time, including: Record the number of tire rotations from the moment the last route sign information was obtained to the current moment as C1; Based on the line marker information, the number of line markers from the time the RFID station signal was last acquired to the current time is recorded as B; Based on the latest RFID station signal acquired at the current moment, determine the station closest to the inspection robot, and combine the number of tire rotations C1 and the number of line markers B to locate the current position of the inspection robot. The current position of the positioning inspection robot is: B*D+C1*L, which is the distance from the Xth station; Where X represents the number of stations corresponding to the latest RFID station signal acquired at the current moment, B represents the number of line markers from the latest RFID station signal acquired to the current moment, D represents the set distance between adjacent line markers, C1 represents the number of tire rotations from the latest time of acquiring line marker information to the current moment, and L represents the wheel circumference of the inspection robot. During the process of recording the number of tire rotations C1 from the moment the latest route sign information was obtained to the current moment, if an RFID station signal is obtained, the currently recorded number of tire rotations C1 is cleared and recorded again. During the comprehensive positioning process, the positioning results are corrected using conflict correction methods, including: If the current distance traveled exceeds the set distance between two route markers, the route marker information acquisition is considered to have failed. The route marker information is automatically supplemented, and the current distance traveled is cleared and re-recorded. The current distance traveled is the product of the number of tire rotations C1 from the time the route marker information was last acquired to the current time and the circumference L of the inspection robot's wheel, i.e., C1*L.

2. The integrated positioning method for subway tunnel inspection robots as described in claim 1, characterized in that it further... include: If the encoder fails to record the number of tire rotations C1 from the moment the last route marker information was acquired to the current moment, the encoder will be restarted and re-recorded after the next route marker information is received.

3. The integrated positioning method for subway tunnel inspection robots as described in claim 1, characterized in that, Record the number of tire rotations from the initial inspection moment to the current moment as C. Combine this with the wheel circumference L of the inspection robot to calculate the total distance traveled by the inspection robot, which is C*L.

4. A comprehensive positioning system for a subway tunnel inspection robot, characterized in that, It includes an inspection robot and a background positioning system; the inspection robot is equipped with a servo motor encoder, inertial sensor, camera and RFID tag; RFID receivers are installed on both sides of each platform in the subway tunnel; and multiple adjacent line markers are evenly distributed on the line between every two platforms in the subway tunnel. The servo motor encoder, inertial sensor, camera, RFID tag and RFID receiver are used to acquire in real time the number of tire rotations of the inspection robot, robot motion parameters, captured tunnel images, and RFID platform signals fed back by the inspection robot when it passes the platform. The inspection robot is equipped with an onboard computer, which is connected to the background positioning system to receive acquired data information, perform preliminary processing on the data information, and transmit the preliminarily processed data information to the background positioning system. The background positioning system is used to perform comprehensive positioning of the inspection robot based on the real-time acquired RFID station signals, tire rotation count, line marking information and pose information, and output the real-time position information and pose information of the inspection robot. The priority of the positioning methods based on servo motor encoders and inertial navigation, the positioning methods based on computer vision to identify tunnel mileage, and the RFID-based interval positioning methods is as follows: RFID-based interval positioning method > Computer vision to identify tunnel mileage positioning method > Servo motor encoder and inertial navigation positioning method. Comprehensive positioning of the inspection robot in real time, including: Record the number of tire rotations from the moment the last route sign information was obtained to the current moment as C1; Based on the line marker information, the number of line markers from the time the RFID station signal was last acquired to the current time is recorded as B; Based on the latest RFID station signal acquired at the current moment, determine the station closest to the inspection robot, and combine the number of tire rotations C1 and the number of line markers B to locate the current position of the inspection robot. The current position of the positioning inspection robot is: B*D+C1*L, which is the distance from the Xth station; Where X represents the number of stations corresponding to the latest RFID station signal acquired at the current moment, B represents the number of line markers from the latest RFID station signal acquired to the current moment, D represents the set distance between adjacent line markers, C1 represents the number of tire rotations from the latest time of acquiring line marker information to the current moment, and L represents the wheel circumference of the inspection robot. During the process of recording the number of tire rotations C1 from the moment the latest route sign information was obtained to the current moment, if an RFID station signal is obtained, the currently recorded number of tire rotations C1 is cleared and recorded again. During the comprehensive positioning process, the positioning results are corrected using conflict correction methods, including: If the current distance traveled exceeds the set distance between two route markers, the route marker information acquisition is considered to have failed. The route marker information is automatically supplemented, and the current distance traveled is cleared and re-recorded. The current distance traveled is the product of the number of tire rotations C1 from the time the route marker information was last acquired to the current time and the circumference L of the inspection robot's wheel, i.e., C1*L.

Images