Remote-controlled monorail defect real-time inspection robot

Abstract

The application relates to the technical field of robots, and discloses a single-track defect real-time inspection robot controlled remotely, which comprises a robot body, a driving wheel and a driving system; the robot body comprises an inspection platform; the driving wheel is arranged at the lower part of the inspection platform and is used for moving along a track; the driving system is used for providing power for the robot; a control device comprises a sensor module and a control component; the sensor module is used for acquiring parameters in the motion process of the robot; the control component is connected with the sensor module, is used for controlling the starting of the driving system, and acquires signals collected by the sensor module; a telescopic mechanical arm is movably connected with the inspection platform at one end and is a distal end at the other end; the telescopic mechanical arm is connected with the driving system and the control component; the application combines the telescopic mechanical arm and the sensor module to acquire more accurate track information; a fast inspection combined with slow inspection method is adopted, so that more accurate inspection results can be obtained.

Description

Technical Field

[0001] This invention relates to the field of robotics, specifically to a remotely controlled single-track real-time defect inspection robot. Background Technology

[0002] Track inspection robots are inspection devices used to inspect track systems such as railways and subways. Existing inspection robots can typically navigate according to a pre-planned inspection path. They can collect and analyze data in real time using various types of sensors, and obtain preliminary status monitoring results by processing and judging the data through existing algorithms and models.

[0003] However, most existing track inspection robots can only perform inspections at fixed angles and positions. Due to the complex and varied terrain, blind spots exist when inspecting at a fixed angle, which is not conducive to comprehensive inspection. Furthermore, existing inspection robots also suffer from problems such as limited sensor types, insufficient detection accuracy, and inability to accurately locate objects. Summary of the Invention

[0004] This invention addresses the problems existing in current inspection robot technology by providing a remotely controlled, single-track, real-time defect inspection robot.

[0005] The technical solution adopted in this invention is: A remotely controlled, single-track, real-time defect inspection robot, comprising: The robot body includes an inspection platform, drive wheels, and a drive system; the inspection platform moves along a track via the drive wheels located at its lower part; the drive system provides power to the robot. The control device includes a sensor module and a control component. The sensor module is used to acquire parameters during the robot's movement. The control component is connected to the sensor module and is used to control the start of the drive system and acquire the signals collected by the sensor module. The telescopic robotic arm has one end connected to the inspection platform and the other end remotely; the telescopic robotic arm is connected to the drive system and control components. The control process of the control component is as follows: First, start the drive system and adjust the robot speed. The robot then quickly inspects the pre-set detection area to identify areas that may have faults or damage. Then the control drive system is started, the robot travels to the area where there may be faults and damage, the robot speed is adjusted, and a slow inspection is carried out.

[0006] Furthermore, the sensor module includes an accelerometer sensor mounted on the inspection platform for acquiring the robot's acceleration; It also includes an image acquisition device for acquiring image information during the inspection process, and the image acquisition device is connected to the control component; The control component acquires data collected by the accelerometer and obtains an accelerometer data sequence. Based on cross-correlation, the SBD distance of the obtained accelerometer data sequence is calculated; Calculate the centroid of the accelerometer data sequence based on the SBD distance; Clustering is performed based on the obtained centroids to identify suspicious regions; Extract the time point of the suspicious area and find the intersection of it with the image data obtained at that time point; Identify areas that may have malfunctions or damage.

[0007] Furthermore, the sensor module also includes a TMR sensor and a laser displacement sensor disposed at the distal end of the telescopic robotic arm; The control components acquire information about defects or damage inside the track based on signals measured by the TMR sensors; The control components acquire damage information about the outside of the track based on signals measured by the laser displacement sensor.

[0008] Furthermore, the telescopic robotic arm includes a first robotic arm and a second robotic arm that are movably connected; the first robotic arm is movably connected to the inspection platform; both the first robotic arm and the second robotic arm are telescopic structures; the control component controls the rotation between the first robotic arm and the inspection platform, and between the first robotic arm and the second robotic arm; the control component controls the extension and retraction of the first robotic arm and the second robotic arm.

[0009] Furthermore, a sensor bracket is provided at the end of the second robotic arm away from the first robotic arm, and is movably connected thereto; the control component controls the movement between the sensor bracket and the second robotic arm; Both the TMR sensor and the laser displacement sensor are mounted on the sensor bracket.

[0010] Furthermore, the sensor module also includes a distance measuring sensor mounted on the sensor bracket; the distance measuring sensor is used to measure the distance between the sensor bracket and the track.

[0011] Furthermore, the control component includes a remote control component and a control board; the remote control component and the control board are connected via wireless transmission. It also includes a GPS positioning device for obtaining the location information of the inspection robot, and the GPS positioning device is electrically connected to the control board.

[0012] Furthermore, the drive system includes a motor connected to an electromagnetic brake; it also includes a transmission device connected to the output shaft of the motor; the transmission device provides power to the drive wheel via a belt; the motor is connected to a battery module; and it also includes guide wheels disposed on both sides of the track for guiding the inspection platform.

[0013] Furthermore, the first robotic arm and the second robotic arm are connected by a first joint; the second robotic arm and the sensor bracket are connected by a second joint.

[0014] The beneficial effects of this invention are: (1) The present invention adopts a method of combining rapid inspection with slow inspection, which can obtain more accurate inspection results; (2) The present invention is equipped with a telescopic robotic arm, the length and angle of which can be adjusted, and the sensor bracket on which it is mounted can be adjusted in angle and posture to meet different inspection needs; (3) The present invention is equipped with a sensor module, including a TMR sensor, a laser displacement sensor, an acceleration sensor, and a distance sensor, which can acquire different signals. Attached Figure Description

[0015] Figure 1 This is a rear-view stereoscopic structural diagram of the inspection robot of the present invention.

[0016] Figure 2 This is a front-view stereoscopic structural diagram of the inspection robot of the present invention.

[0017] Figure 3 This is a side view of the inspection robot of the present invention.

[0018] Figure 4 This is a front view of the inspection robot of the present invention.

[0019] Figure 5 This is a top view of the inspection robot of the present invention.

[0020] Figure 6 This is a schematic diagram of the internal structure of the inspection robot body of the present invention.

[0021] Figure 7 This is a schematic diagram of the drive system in the inspection robot of the present invention.

[0022] Figure 8 This is a schematic diagram of the transmission structure of the drive system in the inspection robot of the present invention.

[0023] Figure 9 This is a schematic diagram of the posture of the inspection robot of the present invention during rapid inspection.

[0024] Figure 10 This is a schematic diagram of the slow-speed inspection posture of the inspection robot of the present invention.

[0025] In the diagram: 1-track, 2-drive wheel, 3-accelerometer, 4-wireless signal receiver, 5-guide wheel, 6-bearing, 7-sensor bracket, 8-first joint, 9-robotic arm mounting base, 10-second joint, 11-inspection platform, 12-first camera, 13-first robotic arm, 14-second robotic arm, 15-halogen lamp, 16-TMR sensor, 17-range sensor, 18-laser displacement sensor, 19-second camera, 20-GPS positioning device, 21-control board, 22-wireless signal transmitter, 23-electromagnetic brake, 24-motor, 25-transmission device, 26-battery module, 27-belt. Detailed Implementation

[0026] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0027] A remotely controlled, single-track, real-time defect inspection robot includes: The robot body includes an inspection platform 11, drive wheels, and a drive system; the inspection platform moves along the track 1 via drive wheels 2 located at its lower part; the drive system provides power to the robot. The control device includes a sensor module and a control component. The sensor module is used to acquire parameters during the robot's movement. The control component is connected to the sensor module and is used to control the start of the drive system and acquire the signals collected by the sensor module. The telescopic robotic arm has one end connected to the inspection platform and the other end remotely; the telescopic robotic arm is connected to the drive system and control components. The sensor module also includes a TMR sensor 16 and a laser displacement sensor 18 located at the far end of the telescopic robotic arm; The control component acquires information about defects or damage inside track 1 based on signals measured by TMR sensor 16. TMR sensor 16 characterizes defects or cracks inside track 1 by detecting physical information such as the magnitude, direction, displacement, angle, and current of the magnetic field applied to track 1. The signal processing methods and judgment methods are all achievable with existing technologies, and can be implemented using existing TMR sensor 16 in conjunction with existing technologies, so they will not be elaborated here.

[0028] The control component acquires damage information about the exterior of track 1 based on the signal measured by laser displacement sensor 18. Laser displacement sensor 18 is a 2D digital laser displacement sensor, based on the principle of laser triangulation, projecting a laser beam onto the surface of the object being measured along a line to collect corresponding data. The 2D digital laser displacement sensor integrates a high-performance CMOS receiver, capable of collecting data from thousands of measurement points. The data acquisition unit has analog and digital data output modes, is compatible with other devices, and can be integrated with wireless signal transmitter 22 to transmit data to a remote control component (PC). The inspection robot can use the 2D digital laser displacement sensor to emit a measuring laser onto the track surface. Through laser triangulation and tilt analysis, it can measure the track profile, calculate the inner rail distance, rail wear, left / right rail irregularities, directional unevenness, rail corrugation, and track twisting, among other external track damage information. The judgment method can be achieved using existing technology and will not be elaborated further.

[0029] like Figure 1 , Figure 2 , Figure 3 and Figure 4 As shown, the telescopic robotic arm includes a first robotic arm 13 and a second robotic arm 14 that are movably connected. The first robotic arm 13 is movably connected to the inspection platform 11. Both the first robotic arm 13 and the second robotic arm 14 are telescopic structures. A control component controls the rotation between the first robotic arm 13 and the inspection platform 11, and between the first robotic arm 13 and the second robotic arm 14. The control component also controls the extension and retraction of the first robotic arm 13 and the second robotic arm 14. The first robotic arm 13 and the inspection platform 11 can rotate relative to each other, which can be achieved by setting an axis. Here, the control component can control the first robotic arm 13 to rotate around the robotic arm mounting base 9, and lock it in place by a latch or other means when it reaches a set position. This method can be achieved through various structures, or it can be achieved manually.

[0030] The first robotic arm 13 includes two interlocking mechanical rods. The first mechanical rod can extend and retract along the second mechanical rod, which can be achieved through a hydraulic device. The extension and retraction amount is controlled by a pre-set distance, and the arm locks at the set position. The control component controls the extension and retraction amount by controlling the opening and closing of the hydraulic device. The second robotic arm 14 has the same structure as the first robotic arm 13. Of course, other extendable devices can also be used to achieve the same purpose. There are various possible structures in the prior art.

[0031] The second robotic arm 14 is provided with a sensor bracket 7 that is movably connected to the end of the first robotic arm 13; the control component controls the movement between the sensor bracket 7 and the second robotic arm 14; the first robotic arm 13 and the second robotic arm 14 are connected by a first joint 8; the second robotic arm 14 and the sensor bracket 7 are connected by a second joint 10.

[0032] The first joint 8 and the second joint 10 can be configured as a universal joint or similar structure capable of rotation. This type of joint allows for angle adjustment of the robotic arm; it also allows for angle adjustment between the sensor support 7 and the second robotic arm 14. This enables the sensor module to achieve better data acquisition efficiency and results.

[0033] Both the TMR sensor 16 and the laser displacement sensor 18 are mounted on the sensor bracket 7.

[0034] The sensor module also includes a distance sensor 17 mounted on the sensor bracket 7; the distance sensor 17 is used to measure the distance between the sensor bracket 7 and the track 1. This distance information allows for adjustment of the attitude of the sensor bracket 7.

[0035] The control components include a remote control component and a control board 21; the remote control component and the control board 21 are connected wirelessly; the information collected by the control board 21 can be transmitted to the remote control component through a wireless signal transmitter 22; the signals emitted by the remote control component are received by a wireless signal receiver 4 and transmitted to the control board 21 to realize the control of the inspection robot.

[0036] like Figure 5 As shown, it also includes a GPS positioning device 20 for acquiring the location information of the inspection robot, and the GPS positioning device 20 is electrically connected to the control board 21. The GPS positioning device 20 is used to acquire the robot's location information and plan real-time navigation according to the preset inspection path.

[0037] like Figure 5 , Figure 6 , Figure 7 and Figure 8 As shown, the drive system includes a motor 24 connected to an electromagnetic brake 23, which can brake the motor 24 in an emergency, providing a safe braking guarantee for the inspection robot. It also includes a transmission device 25 connected to the output shaft of the motor 24; the transmission device 25 is connected to the drive wheel 2 via a belt 27 to provide power; the motor 24 is connected to a battery module 26, which provides power to the inspection robot; and guide wheels 5 are located on both sides of the track 1 to guide the inspection platform. The drive wheels 2 are drive wheel sets, connected to each other by bearings 6. The guide wheels 5 provide guidance for the inspection robot, ensuring that the inspection robot does not leave the track 1 during operation.

[0038] The control process of the control component is as follows: First, start the drive system and adjust the robot speed. The robot then quickly inspects the pre-set detection area to identify areas that may have faults or damage. Then the control drive system is started, the robot travels to the area where there may be faults and damage, the robot speed is adjusted, and a slow inspection is carried out.

[0039] The sensor module includes an accelerometer 3 mounted on the inspection platform to acquire the robot's acceleration; It also includes an image acquisition device for acquiring image information during the inspection process, and the image acquisition device is connected to the control component; The control component acquires the data collected by the accelerometer 3 and obtains the accelerometer data sequence. Based on cross-correlation, the SBD (Shape-Based-Distance) distance of the obtained accelerometer data sequence is calculated; Calculate the centroid of the accelerometer data sequence based on the SBD distance; Clustering is performed based on the obtained centroids to identify suspicious regions; Extract the time point of the suspicious area and find the intersection of it with the image data obtained at that time point; Identify areas that may have malfunctions or damage.

[0040] The image acquisition device includes a first camera 12 mounted on the upper part of the inspection platform 11 and a second camera 19 mounted on the side of the inspection platform 11 near the extendable robotic arm. The first camera 12 is a spherical camera used for obstacle detection. It also includes a halogen lamp 15 for supplemental lighting. This provides illumination to the track when the inspection robot is in a dark or poorly lit environment.

[0041] The specific inspection process of the inspection robot is as follows: The remote control component issues an inspection command, and the wireless signal receiver 4 receives the inspection command and transmits the information to the control board 21.

[0042] Control panel 21 controls the telescopic robotic arm from its initial position. P 0 Adjusted to rapid inspection posture P q ,like Figure 9 As shown.

[0043] (1) in, P 0 represents the initial posture of the retractable robotic arm. l 10 , l 20 These are the initial lengths of the first robotic arm 13 and the second robotic arm 14, respectively. θ 10 , θ 20 , θ 30 , θ 40 These are the relative angles between the robotic arm mounting base 9 and the first robotic arm 13, the first robotic arm 13 and the second robotic arm 14, and the second robotic arm 14 and the sensor bracket 7 in their initial states. P q This is for the rapid inspection posture of the extendable robotic arm. l 1q , l 2q These are the lengths of the first robotic arm 13 and the second robotic arm 14 during the rapid inspection process, respectively. θ 1q , θ 2q , θ 3q , θ 4q These are the relative angles between the robotic arm mounting base 9 and the first robotic arm 13, the first robotic arm 13 and the second robotic arm 14, and the second robotic arm 14 and the sensor bracket 7 in the rapid inspection state. The inspection robot performs inspections in a rapid inspection posture according to instructions, and quickly collects track information using accelerometer 3 and image acquisition device.

[0044] The set of data collected by the accelerometer is represented as , X i The first data collected by the accelerometer i A set of data. x mi For the first i The first group of data sets m Data.

[0045] The image data acquired by the image acquisition device is represented as follows: (2) In the formula: M i The first image acquired by the image acquisition device i Group of two-dimensional image data, For the i-th group of two-dimensional image data, the first... m Line number n The pixel data of the column. The data collected by the first camera and the second camera can both be represented in the form of equation (2).

[0046] The collected data were analyzed based on cross-correlation. For a set of data sequences Sequence translation can be expressed as: (3) in, for m A data sequence consisting of groups of data, This sequence is obtained by shifting the measurement data from the accelerometer. s The data movement step size, ; a shift to the left of the sequence data is negative. Moving to the right is positive. Calculate the inner product between the shifted data sequences to reflect the correlation between the two sets of data.

[0047] Calculate the cross-correlation coefficient of the accelerometer data: (4) (5) In the formula: The cross-correlation coefficient of the acceleration signal sequence after translation. for and The inner product of. A higher value indicates a greater degree of similarity. and The first i Group and No. j The data sequence is obtained by shifting and processing the accelerometer measurement data.

[0048] w For cross-correlation analysis and All possible movements. Due to the acquired data sequence. The number of data inside is m One, then w The range of values ​​is within . k This represents the step size of the data sequence, i.e. .Will As the objective function, the maximum value at w This is the optimal value.

[0049] The SBD distance of the collected data is calculated based on the cross-correlation, and the calculation formula is as shown in equation (6): (6) In the formula: for and SBD distance, for Normalization processing.

[0050] The shape centroids of all data sequences are calculated based on the SBD distance. To avoid the problem that the centroids cannot effectively utilize class features, the centroid calculation is transformed into an optimization problem: finding the minimum sum of the squared distances of all other time series.

[0051] (7) In the formula: For virtual centroid, The sum of squared distances from different sequences of data collected by the accelerometer to the virtual centroid; The centroid that minimizes the sum of squared distances. p k Clusters are formed by clustering.

[0052] Find the minimum distance The problem is transformed into finding its maximum value, which characterizes the squared similarity of the data sequences, as follows: (8) (9) By maximizing the Rayleigh quotient, equation (7) is transformed into equation (8), where Let represent the regularized matrix, and Q = I - O / m , I It is the identity matrix. O It is a matrix consisting entirely of 1s.

[0053] in: , ; In the formula, express transpose; express The segment most similar to the center of mass; express transpose; yes and its transpose The matrix obtained by multiplication is used to simplify calculations. Indicates according to and The regularization matrix is ​​constructed so that it can be simplified using the Rayleigh quotient.

[0054] right M Extracting the first feature vector yields C k The maximum value, i.e. ; can be represented as: .

[0055] Based on the calculated centroids, cluster analysis is performed to cluster the acceleration sensor data from different sequences into a single group. The process is as follows: First, the centroid of the sequence data is calculated and each data set is compared to the centroid. Then, each data set is assigned to the cluster closest to the centroid based on distance. The data within each cluster is updated, and the centroid is updated based on the updated cluster. This assignment process is repeated until the data within each cluster no longer changes. This yields a set of processed accelerometer data. ,right The mean, variance, peak value, and kurtosis of the internal data were analyzed to identify potential damage areas, as follows: (10) In the formula: φ This refers to the area where the track may be damaged. L 0 represents the initial position of the inspection robot on the track. v To improve the speed of the inspection robot, For the mean ,variance Peak x p and kurtosis x q The time point at which damage may occur after feature extraction.

[0056] according to φ Find the image data at the corresponding time point. M if The possibility of damage is cross-validated based on image data.

[0057] (11) In the formula: To verify the potential areas of damage in the track, To φ The intersection of the image data within the range is taken. Taking the intersection can preserve the regions where both the acceleration signal data and the image data are damaged.

[0058] Based on the potential track damage range calculated above, a second slow inspection is conducted.

[0059] The control unit adjusts the inspection robot to a slow inspection posture. P s Slow inspection posture Figure 10 As shown.

[0060] (12) l 1s , l 2sThese are the initial lengths of the first robotic arm 13 and the second robotic arm 14 during slow-speed inspection, respectively. θ 1s , θ 2s , θ 3s , θ 4s These are the relative angles between the robotic arm mounting base 9 and the first robotic arm 13, the first robotic arm 13 and the second robotic arm 14, and the second robotic arm 14 and the sensor bracket 7 in the initial state of slow inspection.

[0061] (13) In the formula: H s The height between the robotic arm mounting base and the track in a slow inspection posture; h s The height between the sensor bracket and the track in slow inspection posture can be obtained by measuring with a laser rangefinder. θ c This represents the angle between the sensor support and the sensor in the slow inspection posture. The slow inspection posture is calculated based on the parameters mentioned above. Indicates about H s , h s and θ c Implicit functions.

[0062] A remote control component controls an inspection robot to perform slow-speed inspections in areas where faults and damage may exist. A tunnel magnetoresistive (TMR) sensor is used to detect the track. The magnetic signal of the magnetoresistive effect within the track area is measured. Convert it into an electrical signal. The electrical signal is transmitted wirelessly to the remote control component for processing. The processing procedure is as follows: (14) In the formula: This indicates that the electrical signal is processed and analyzed, and the magnitude of the magnetic field can be obtained from the electrical signal. Magnetic field angle displacement and current magnitude .

[0063] A 2D digital laser displacement sensor is used to detect the track. A measuring laser is emitted onto the track surface, and the light signal is received by a CMOS receiver built into the sensor. And convert it into an electrical signal .electric signal The signal is transmitted wirelessly to a remote control component for processing. The calculation process is as follows: (15) In the formula: To process and analyze electrical signals, a 2D digital laser displacement sensor uses laser triangulation and tilt analysis to determine the receiving time. Scan length ,inclination and laser energy The process involves measuring the track profile, calculating the inner distance of the rails, rail wear, left-right and directional unevenness, rail corrugation, and track twisting, among other track surface damage.

[0064] The track is slowly inspected using an image acquisition device to collect more detailed data. The acquired two-dimensional image data is represented as follows: (16) In the formula: The first data collected during slow-speed inspection by the camera i Group of two-dimensional image data, For slow inspection i The first group of two-dimensional image data m Line number n The pixel data of the column. The images captured by the first camera and the second camera can both be represented in the form of equation (16).

[0065] Feature-level fusion is performed on the data collected by the TMR sensor, 2D digital laser displacement sensor, and cameras (including the first and second cameras), as follows: (17) In the formula: F The data features are obtained by fusing the data from the three types of sensors. f T For the data characteristics of TMR sensors, f J For the data characteristics of a 2D digital laser displacement sensor, f M The data features from the cameras are used for fault identification based on the fused data features, and inspection data and track status are transmitted in real time to enable staff to handle track defects.

[0066] This invention issues inspection commands via a remote control component, which transmits the signals to the inspection robot's control board via a wireless signal receiver. The control board then controls the operation of other connected components. This allows for control of the telescopic robotic arm's length and angle, as well as the sensor bracket's angle and posture, to meet inspection requirements. The robot's movement is controlled by adjusting the motor's speed, steering, and drive wheel rotation.

[0067] The system employs a combination of rapid and slow inspections, utilizing data from multiple sensors to process and analyze the collected information, thereby enabling accurate fault identification and prediction. Data exchange with remote control components is achieved through wireless signal receivers and transmitters, transmitting inspection data and status information in real time. Statistical analysis of the data is performed (including storage and real-time graphical display using existing software), allowing staff to address track defects promptly.

Claims

1. A remotely controlled monorail real-time defect inspection robot, characterized in that, include: The robot body includes an inspection platform (11), drive wheels and a drive system; the inspection platform (11) moves along the track (1) via drive wheels (2) located at its lower part; the drive system is used to provide power to the robot; The control device includes a sensor module and a control component. The sensor module is used to acquire parameters during the robot's movement. The control component is connected to the sensor module and is used to control the start of the drive system and acquire signals collected by the sensor module. The sensor module includes an accelerometer (3) set on the inspection platform (11) for acquiring the robot's acceleration. It also includes an image acquisition device for acquiring image information during the inspection process, and the image acquisition device is connected to the control component; The control component acquires the data collected by the accelerometer (3) and obtains the accelerometer data sequence; Based on cross-correlation, the SBD distance of the obtained accelerometer data sequence is calculated; the calculation method is as follows: In the formula: and The first i Group and No. j Group acceleration data sequence, The cross-correlation coefficient of the acceleration signal sequence after translation. for and SBD distance, for Normalization processing; Calculate the centroid of the accelerometer data sequence based on the SBD distance; Clustering is performed based on the obtained centroids to identify suspicious regions; Extract the time point of the suspicious area and find the intersection of it with the image data obtained at that time point; Identify areas that may contain faults and damage; The telescopic robotic arm is connected at one end to the inspection platform (11) and at the other end to the remote end; the telescopic robotic arm is connected to the drive system and control components. The control process of the control component is as follows: First, start the drive system and adjust the robot speed. The robot then quickly inspects the pre-set detection area to identify areas that may have faults or damage. Then the control drive system is started, the robot travels to the area where there may be faults and damage, the robot speed is adjusted, and a slow inspection is carried out.

2. The remotely controlled single-track real-time defect inspection robot according to claim 1, characterized in that, The sensor module also includes a TMR sensor (16) and a laser displacement sensor (18) disposed at the far end of the telescopic robotic arm. The control component obtains information on defects or damage inside the track (1) based on the signal measured by the TMR sensor (16); The control component obtains damage information outside the track (1) based on the signal measured by the laser displacement sensor (18).

3. The remotely controlled single-track real-time defect inspection robot according to claim 2, characterized in that, The telescopic robotic arm includes a first robotic arm (13) and a second robotic arm (14) that are movably connected; the first robotic arm (13) is movably connected to the inspection platform (11); both the first robotic arm (13) and the second robotic arm (14) are telescopic structures; the control component controls the rotation between the first robotic arm (13) and the inspection platform (11), and between the first robotic arm (13) and the second robotic arm (14); the control component controls the extension and retraction of the first robotic arm (13) and the second robotic arm (14).

4. The remotely controlled single-track real-time defect inspection robot according to claim 3, characterized in that, The second robotic arm (14) has a sensor bracket (7) that is movably connected to the end of the first robotic arm (13); the control component controls the movement between the sensor bracket (7) and the second robotic arm (14); Both the TMR sensor (16) and the laser displacement sensor (18) are mounted on the sensor bracket (7).

5. A remotely controlled single-track real-time defect inspection robot according to claim 4, characterized in that, The sensor module also includes a distance sensor (17) mounted on the sensor bracket (7); the distance sensor (17) is used to measure the distance between the sensor bracket (7) and the track (1).

6. A remotely controlled single-track real-time defect inspection robot according to claim 5, characterized in that, The control component includes a remote control component and a control board (21); the remote control component and the control board (21) are connected via wireless transmission. It also includes a GPS positioning device (20) for obtaining the location information of the inspection robot, and the GPS positioning device (20) is electrically connected to the control board (21).

7. A remotely controlled single-track real-time defect inspection robot according to claim 6, characterized in that, The drive system includes a motor (24) connected to an electromagnetic brake (23); it also includes a transmission device (25) connected to the output shaft of the motor (24); the transmission device (25) is connected to the drive wheel (2) via a belt (27) to provide power to it; the motor (24) is connected to a battery module (26); and it also includes guide wheels (5) set on both sides of the track (1) for guiding the inspection platform (11).

8. A remotely controlled single-track real-time defect inspection robot according to claim 7, characterized in that, The first robotic arm (13) and the second robotic arm (14) are connected by a first joint (8); the second robotic arm (14) and the sensor bracket (7) are connected by a second joint (10).

Images