Subway inspection robot cooperative operation method and system

Abstract

The application belongs to the technical field of intelligent detection of rail transit, and particularly relates to a subway inspection robot cooperative operation method and system. The method comprises the following steps: setting a first preset starting point, a second preset starting point and a first preset ending point according to the bogie position of a vehicle to be inspected; controlling a first robot and a second robot to enter the corresponding trench of the vehicle to be inspected according to an inspection task, and respectively reaching the first preset starting point and the second preset starting point; after reaching the preset starting point, controlling the first robot to move to the bogie area of the current vehicle and to perform relative positioning with the current vehicle; controlling the relative distance between the second robot and the first robot to be the preset distance; after the first robot completes the relative positioning, controlling the first robot and the second robot to respectively perform image acquisition on the underfloor components of the current vehicle; and when the first robot reaches the first preset ending point, controlling the second robot to close the follow-up mode.

Description

Technical Field

[0001] This application belongs to the field of intelligent detection technology for rail transit, specifically relating to a collaborative operation method and system for subway inspection robots. Background Technology

[0002] After subway trains finish operation, they must return to the depot for safety inspections to ensure there are no missing, damaged, cracked, or leaking parts. Current inspection methods primarily rely on manual visual inspection, which is labor-intensive and prone to omissions and errors.

[0003] A few subway companies have begun to explore the use of undercarriage robots for inspection, but the existing robot solutions have the following problems: First, the efficiency of single-machine operation is low, and each carriage needs to be stopped and positioned multiple times, making it difficult to complete the inspection of the entire train during the off-season; second, it is impossible to achieve multi-robot collaborative operation, resulting in low overall inspection efficiency.

[0004] Therefore, there is an urgent need for a subway inspection method that can enable collaborative robot operations and improve inspection efficiency. Summary of the Invention

[0005] This invention solves at least to some extent the above-mentioned technical problems and provides a method and system for collaborative operation of subway inspection robots. By presetting the starting point and maintaining the wheelbase, the front robot can be positioned once and the rear robot can automatically land and collect data synchronously, thereby improving inspection efficiency.

[0006] The first aspect of the embodiments of this disclosure provides a method for collaborative operation of subway inspection robots, including: Based on the bogie position of the vehicle to be inspected, a first preset starting point, a second preset starting point, and a first preset ending point are set, and the distance between the two preset starting points in the trench direction is the preset distance. According to the inspection task, the first robot and the second robot are controlled to enter the pit corresponding to the vehicle to be inspected and reach the first preset starting point and the second preset starting point respectively. After reaching the preset starting point, the first robot is controlled to move to the bogie area of ​​the current carriage and perform relative positioning with the current carriage; the second robot is controlled to start the follow-up mode, in which the relative distance between the second robot and the first robot is maintained at the preset distance. After the first robot completes its relative positioning with the current carriage, it controls the first robot and the second robot to respectively acquire images of the undercarriage components of the current carriage; When the first robot reaches the first preset endpoint, the second robot is controlled to turn off the follow-up mode.

[0007] The technical solution provided in this application brings at least the following benefits: by setting a preset starting distance and controlling the follow-up mode, the synchronous operation of two robots can be realized, reducing the number of relative positioning operations and improving inspection efficiency.

[0008] In other embodiments of this application, the method further includes: After the first robot and the second robot reach the first preset starting point and the second preset starting point respectively, the lateral distance between the first robot and the second robot and the walls on both sides of the trench is kept consistent so that the first robot and the second robot walk along the center line of the trench.

[0009] The technical solution provided in this application brings at least the following benefits: ensuring that the two robots walk along the centerline in the narrow trench, avoiding collisions with the side walls, and ensuring driving safety.

[0010] In other embodiments of this application, the preset distance is the wheelbase of the bogie of the vehicle to be inspected.

[0011] The technical solution provided in this application brings at least the following benefits: by setting the preset distance as the bogie wheelbase, two robots are aligned with the front and rear axles of the same bogie, achieving synchronous data acquisition.

[0012] In other embodiments of this application, the method further includes: After the second robot closes the follow-up mode, it controls the first robot and the second robot to drive out of the ditch and drive along both sides of the vehicle to be repaired toward the preset charging point. During the journey to the preset charging point, the first robot and the second robot are controlled to capture images of the visible components on one side of each robot.

[0013] The technical solution provided in this application brings at least the following beneficial effects: by diverting the robot to both sides of the vehicle after exiting the ditch, and completing the vehicle-side inspection during the return trip, multiple tasks can be completed in a single trip. In other embodiments of this application, after the first robot and the second robot drive out of the trench, a driving path to their respective preset charging points is planned based on the current positions of the first robot and the second robot and their respective preset charging points.

[0014] The technical solution provided in this application brings at least the following beneficial effects: by dynamically planning the route based on the current location and charging point, path intersections and duplicate detections are avoided, thereby improving return efficiency. In other embodiments of this application, according to the travel path, the first robot and the second robot travel along one side of the vehicle to be repaired toward their respective preset charging points in opposite directions.

[0015] The technical solution provided in this application brings at least the following beneficial effects: by placing the two vehicles on opposite sides of the vehicle and driving in opposite directions, mutual interference is avoided, and full coverage detection of the vehicle side is achieved.

[0016] In other embodiments of this application, during the journey to a preset charging point, the first robot and the second robot are controlled to collect infrared thermal imaging data of the environment through their respective infrared cameras. The infrared thermal imaging data is analyzed in real time. When the temperature of any area exceeds a preset threshold, it is determined to be a temperature abnormality event. Based on the temperature abnormality event, the system triggers an alarm prompt.

[0017] The technical solution provided in this application brings at least the following beneficial effects: by monitoring the ambient temperature while detecting on the vehicle side, multiple tasks can be performed concurrently, and potential safety hazards can be detected in a timely manner.

[0018] In other embodiments of this application, the second robot is controlled to activate a follow-up mode, wherein the relative distance between the second robot and the first robot is maintained at the preset distance in the follow-up mode, including: The second robot detects its distance from the first robot using a multi-line positioning lidar. Based on its distance from the first robot, the second robot calculates its compensation value relative to the first robot in the trench direction; Based on the compensation value, the second robot accelerates or decelerates to maintain the relative distance at the preset distance.

[0019] The technical solution provided in this application brings at least the following beneficial effects: by utilizing lidar for real-time ranging and dynamic compensation, high-precision distance-keeping control is achieved.

[0020] In other embodiments of this application, the relative positioning with respect to the current carriage includes: The first robot is controlled to acquire an image of the bogie area of ​​the current carriage through the top camera. The acquired image of the bogie area of ​​the current carriage is matched and identified with a pre-stored bogie center pin template to obtain the pixel coordinates of the bogie center pin in the image. Based on the pixel coordinates and the pre-stored installation height of the top camera, the relative pose deviation between the first robot and the center pin of the bogie is calculated through coordinate transformation; Based on the positional deviation, the first robot is controlled to make micro-adjustments, so that the first robot moves directly below the pre-stored shooting point of the bogie center pin.

[0021] The technical solution provided in this application brings at least the following benefits: by visually recognizing the center pin and calculating the pose deviation, precise positioning of the robot and the vehicle can be achieved.

[0022] A second aspect of this disclosure provides a collaborative operation system for subway inspection robots, used to implement the collaborative operation method for subway inspection robots disclosed in the first aspect of this application. The system includes: The platform system is configured to issue control commands according to the inspection task, and set a first preset starting point, a second preset starting point, and a first preset ending point according to the carriage of the vehicle to be inspected. The distance between the two preset starting points in the direction of the trench is a preset distance. Both the first and second robots are equipped with top-mounted cameras on their upper surfaces to capture key feature points of the underside of the vehicle to be inspected. The second robot includes a multi-line positioning LiDAR for real-time detection of the relative distance to the first robot. The first and second robots communicate wirelessly. The first and second robots are configured as follows: According to the system control instructions, the vehicle enters the pit corresponding to the vehicle to be detected and arrives at the first preset starting point and the second preset starting point respectively; Upon reaching the preset starting point, the first robot moves to the bogie area of ​​the current carriage according to the system control command and performs relative positioning with the current carriage; the second robot starts the follow-up mode, in which the relative distance between the second robot and the first robot is maintained at the preset distance; After the first robot completes its relative positioning with the current carriage, the first robot and the second robot respectively acquire images of the undercarriage components of the current carriage according to the system control instructions; When the first robot reaches the first preset endpoint, the second robot terminates the follow-up mode according to the system control command. Compared with the prior art, the present invention has the following beneficial effects: (1) By presetting the starting point and wheelbase, a synchronous acquisition mechanism is realized in which the front car is positioned once and the rear car is automatically placed on the rear axle, which reduces the number of positioning times for each bogie by half, significantly improves inspection efficiency, and meets the needs of shortening the subway operation window.

[0023] (2) After exiting the ditch, the two robots automatically diverge to the sides of the vehicle and complete vehicle-side detection and environmental monitoring in parallel on the return journey, enabling multiple tasks to be completed in one trip, avoiding repeated paths and improving the integration of operations.

[0024] The above description is merely an overview of the technical solution disclosed herein. In order to better understand the technical means of this disclosure and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this disclosure more apparent and understandable, specific embodiments of this disclosure are described below. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is the flowchart of the collaborative operation method of subway inspection robots in the embodiments of this application. Figure 1 ; Figure 2 This is a schematic diagram of a collaborative operation method for subway inspection robots according to an embodiment of this application. Figure 1 ; Figure 3 This is the flowchart of the collaborative operation method of subway inspection robots in the embodiments of this application. Figure 2 ; Figure 4 This is the flowchart of the collaborative operation method of subway inspection robots in the embodiments of this application. Figure 3 ; Figure 5 This is a topology diagram of the collaborative operation system of the subway inspection robot according to an embodiment of this application; Figure 6 This is a schematic diagram of the subway inspection robot structure according to an embodiment of this application. Figure 1 ; Figure 7 This is a schematic diagram of the subway inspection robot structure according to an embodiment of this application. Figure 2 ; Figure 8 This is a schematic diagram of a collaborative operation method for subway inspection robots according to an embodiment of this application. Figure 2 ; In the above figures, 1. Platform system; 2. First robot; 3. Second robot; 4. Top camera; 5. Multi-line positioning LiDAR; 6. Supplemental lighting source; 7. Vehicle-mounted industrial control computer; 8. Four-wheel drive wheel system; 9. Six-axis robotic arm; 10. Vision camera; 11. Lifting column; 12. Infrared camera; 13. Obstacle detection LiDAR. P1, Marker point; P2, Second preset starting point; P3, First preset starting point; P4, First preset ending point; P5, Second preset ending point. Detailed Implementation

[0027] To make the technical problems, technical solutions, and beneficial effects to be solved by this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and are not intended to limit the scope of this application.

[0028] The prefixes such as "first" and "second" used in this application embodiment are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this application embodiment does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not constitute unnecessary restrictions due to the use of such prefixes. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0029] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B; the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone.

[0030] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0031] In this application, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0032] Current methods for inspecting the exterior of subway vehicles primarily rely on flashlight illumination and manual visual inspection. This requires inspections during off-peak hours, is labor-intensive, and is highly susceptible to human error, increasing the risk of missed or incorrect inspections. A few subway companies have begun exploring a combination of manual inspection and undercarriage robot inspection. However, the robots have limited functionality, only inspecting the undercarriage and partially replacing manual labor. Furthermore, each robot requires precise positioning of every car during individual operation, resulting in low efficiency.

[0033] To address the above problems, this application provides a collaborative operation method and system for subway inspection robots. By pre-setting the starting point and wheelbase, the system achieves one-time positioning of the front robot and automatic placement and synchronous data collection of the rear robot, thereby improving inspection efficiency. The specific implementation method of this application will be described in detail below with reference to the accompanying drawings.

[0034] The first aspect of this disclosure discloses a method for collaborative operation of subway inspection robots, such as... Figure 1 As shown, the method includes: S01: Based on the bogie position of the vehicle to be inspected, set a first preset starting point, a second preset starting point, and a first preset ending point. The distance between the two preset starting points in the trench direction is the preset distance. The first preset starting point P3 and the second preset starting point P2 are respectively set in the trench at positions corresponding to the vicinity of the first set of bogies of the vehicle to be inspected. For example, the first preset starting point P3 is set about 0.5 meters in front of the front axle of the first set of bogies, serving as the entry and stopping point for the first robot; the second preset starting point P2 is set about 0.5 meters behind the rear axle of the first set of bogies, serving as the entry and stopping point for the second robot.

[0035] The two preset starting points are set based on the following: the length of the car body of the vehicle to be inspected determines the overall layout of the bogie; the wheelbase of the bogie determines the relative position between the front and rear axles; and the width of the trench and the location of the ramp determine the initial stopping area of ​​the robot after it enters the trench.

[0036] The distance between the two preset starting points in the direction of the ditch is set as a preset distance.

[0037] The first preset endpoint P5 is the end position where the first robot completes the undercarriage inspection. It is set inside the pit, corresponding to the rear axle of the last set of bogies. For example, it is set about 0.5 meters behind the rear axle of the last set of bogies. The preset endpoint is set based on the following: the number of trains to be inspected and the length of the carriage determine the position of the last set of bogies; the endpoint position of the pit determines the transition area where the robot exits the pit.

[0038] In some other embodiments, the platform system also pre-sets a second preset endpoint P4, which is located within the trench directly below or slightly in front of the rear axle of the last set of bogies. The second preset endpoint P4 serves as the trigger point for the second robot to complete its detection.

[0039] For example, based on the length of the vehicle's cargo box and the bogie layout, the system pre-sets a first preset starting point P3, a second preset starting point P2, and a first preset ending point P5. Figure 2 As shown, the first preset starting point P3 and the second preset starting point P2 are located at the first end of the trench, and the first preset ending point is located at the second end of the trench. According to the system's inspection task, the corresponding vehicles to be inspected begin to be inspected from the first end of the trench. The distance between the first preset starting point P3 and the second preset starting point P2 in the trench direction is a preset distance. The coordinate parameters of the first preset starting point P3, the second preset starting point P2, and the first preset ending point P5 are stored in the platform system as a reference for subsequent robot entry into the trench and collaborative operations.

[0040] S02: According to the inspection task, control the first robot and the second robot to enter the pit corresponding to the vehicle to be inspected, and reach the first preset starting point and the second preset starting point respectively; For example, based on the inspection task, the platform system sends control commands through a wireless router to control the first and second robots to move from the depot floor via a ramp into the maintenance pit corresponding to the vehicle to be inspected. The first and second robots move to the preset first starting point P3 and the preset second starting point P2, respectively, and report their arrival information to the platform system.

[0041] In a specific illustrative embodiment, the platform system also sets a marker point P1 on the ground area in front of the entrance of the first end ramp of the trench, which serves as the trigger point for role assignment before the robot enters the trench. The platform system controls two robots to move towards the vehicle to be inspected according to the inspection task, and sets the robot that arrives at the marker point P1 first as the first robot, and the robot that arrives at the marker point P1 later as the second robot.

[0042] The technical solution of this application achieves automatic role allocation by setting marker points outside the trench, designating the robot that arrives first as the first robot (master) and the robot that arrives later as the second robot (follower). This rule simplifies the scheduling logic, avoids master-slave conflicts before entering the trench, ensures that the two robots enter the trench in a clear order, establishes a clear master-slave relationship for subsequent collaborative operations, and ensures a smooth start to the collaborative process.

[0043] S03: After reaching the preset starting point, control the first robot to move to the bogie area of ​​the current carriage and perform relative positioning with the current carriage; control the second robot to start the follow-up mode. In the follow-up mode, the relative distance between the second robot and the first robot is maintained at the preset distance. Follow-up mode refers to a collaborative working mode in which the second robot uses the first robot as a reference device, detects the relative distance between them in real time, and autonomously adjusts its speed to maintain a preset distance between the two robots. In this mode, the second robot does not rely on the platform system's real-time path planning instructions, but instead uses the first robot's motion state as a reference to autonomously complete the following control, thereby simplifying the collaborative logic and reducing communication overhead.

[0044] For example, after the first robot and the second robot reach the first preset starting point P3 and the second preset starting point P2 respectively, they report their position arrival information to the platform system. The platform system issues a collaborative control command. According to the collaborative control command, the first robot moves to the bogie area of ​​the current car. After reaching the bogie area, the first robot uses the top camera 4 and the supplementary light source 6 installed on the upper surface of its shell to take pictures of key feature points of the subway car bottom, such as the center pin of the bogie, and returns the key feature point images. The platform system analyzes the key feature point images through visual recognition technology to determine the relative position of the first robot with respect to the current car, thereby achieving accurate positioning of the relative position of the first robot with respect to the current car.

[0045] Specifically, in a particular illustrative embodiment, such as Figure 3 , 8 As shown, controlling the first robot to move to the bogie area of ​​the current carriage and perform relative positioning with the current carriage includes: S21: Control the first robot to acquire an image of the bogie area of ​​the current carriage through the top camera, and perform image matching and recognition between the acquired image of the bogie area of ​​the current carriage and the pre-stored bogie center pin template to obtain the pixel coordinates of the bogie center pin in the image. S22: Based on the pixel coordinates and the pre-stored installation height of the top camera, calculate the relative pose deviation between the first robot and the center pin of the bogie through coordinate transformation; S23: Based on the pose deviation, control the first robot to make micro-adjustments so that the first robot moves directly below the pre-stored shooting point of the bogie center pin.

[0046] For example, the top camera 4 of the first robot acquires an image of the current bogie area of ​​the carriage. The onboard industrial control computer 7 of the first robot uses a pre-stored bogie center pin image template to perform feature matching and recognition on the acquired image to obtain the pixel coordinates of the center pin in the image.

[0047] The onboard industrial control computer 7 of the first robot calculates the relative pose deviation between the first robot body and the center pin of the bogie through a coordinate transformation algorithm based on the pixel coordinates and the pre-stored calibration parameters such as the installation height and installation angle of the top camera 4. This includes the X-axis deviation, Y-axis deviation, and angle deviation.

[0048] Based on the calculated pose deviation, the onboard industrial control computer 7 of the first robot sends control commands to its four-wheel drive chassis to control the first robot to perform micro-adjustments. These micro-adjustments include: controlling the four sets of four-wheel drive wheels 8 to simultaneously turn 90° to achieve lateral translation, or controlling the front and rear four sets of four-wheel drive wheels 8 to turn at specific angles to achieve oblique movement, so that the first robot accurately moves to directly below the pre-stored center pin shooting point of the bogie, completing its relative positioning with the vehicle.

[0049] The location of subway cars when parking in the pit varies randomly, and directly relying on preset positions for photography can lead to image distortion and recognition difficulties. This technical solution uses visual recognition of the center pin and calculates the pose deviation, enabling the first robot to accurately calibrate its relative position to the car, eliminating positioning errors caused by the car's parking deviation.

[0050] Specifically, in a particular illustrative embodiment, such as Figure 4 As shown, in follow-up mode, the relative distance between the second robot and the first robot remains at a preset distance, including: S31: The second robot detects the distance to the first robot using a multi-line positioning lidar; S32: Based on its distance from the first robot, the second robot calculates its compensation value relative to the first robot in the trench direction; S33: Based on the compensation value, the second robot accelerates or decelerates to maintain the relative distance at the preset distance.

[0051] For example, the multi-line positioning LiDAR 5 of the second robot continuously scans the surrounding environment to acquire point cloud data including that of the first robot. The on-board industrial control computer 7 of the second robot identifies the position coordinates of the first robot through a point cloud data processing algorithm, and calculates the distance between the center points of the first robot and the second robot in the direction of the trench, as a relative distance detection value.

[0052] The onboard industrial control computer 7 of the second robot compares the detected value with the preset distance and calculates the compensation value. When the detected value is greater than the preset distance, an acceleration command is generated to control the four-wheel drive wheel system 8 of the second robot to increase the travel speed and reduce the distance; when the detected value is less than the preset distance, a deceleration command is generated to reduce the travel speed and increase the distance. Closed-loop control of the distance is achieved through a PID control algorithm, so that the relative distance is always kept within the allowable error range of the preset distance.

[0053] The technical solution of this application uses a multi-line positioning lidar 5 to detect distance in real time and a dynamic compensation mechanism implemented by a PID closed-loop control algorithm. This enables the second robot to maintain a precise relative distance with the first robot without additional positioning, providing position assurance for subsequent synchronous data acquisition.

[0054] In a specific exemplary embodiment, the preset distance is the wheelbase of the bogie of the vehicle to be inspected. After the first robot completes the relative positioning with the current carriage, it moves to the front axle of the bogie of the current carriage according to the instructions of the platform system. Since the distance between the second robot and the first robot is always kept at the wheelbase of the bogie, the second robot moves with the first robot to the rear axle of the bogie of the current carriage. The platform system controls the first robot and the second robot to simultaneously start the vision cameras at the end of their respective robotic arms, respectively to collect images of the components at the target acquisition point, and store the collected images in their respective on-board industrial control computers.

[0055] The technical solution of this application presets the starting points of the two robots entering the ditch to be one bogie wheelbase apart, so that the two robots naturally form a front-to-back position relationship after entering the ditch. When the first robot completes its positioning and moves to the front axle, the second robot can automatically be positioned at the rear axle without additional positioning, laying the positional foundation for subsequent synchronous data acquisition and avoiding the time consumption of stopping and positioning twice for each carriage in the traditional solution.

[0056] S04: After the first robot completes its relative positioning with the current carriage, control the first robot and the second robot to collect images of the undercarriage components of the current carriage respectively; Both the first and second robots are equipped with complete image acquisition equipment, including: a lifting column 11, a six-axis robotic arm 9, a robotic arm end vision camera 10, and a top camera 4.

[0057] The lifting column 11 is mounted on the robot chassis and can be raised and lowered vertically to adjust the overall working height of the six-axis robotic arm 9, adapting to the underbody height of different vehicle models and the depth of maintenance pits. The six-axis robotic arm 9 is bolted to the top of the lifting column 11 and has six degrees of freedom of flexible movement, enabling multi-angle and multi-pose image acquisition. The end effector vision camera 10 is fixed to the end of the six-axis robotic arm 9 via mounting hardware and is used to acquire detailed images of key components under the vehicle. The top camera 4 is mounted on the upper surface of the robot's outer shell and is used for scanning and acquiring global images of the underbody.

[0058] For example, after the first robot completes its relative positioning with the current carriage, it moves to directly below the target acquisition point assigned to it by the system according to the platform system instructions. Based on the pre-stored coordinates of the front axle photography point of the bogie, the first robot controls the lifting column 11 to rise to a preset height, so that the end-effector vision camera 10 of the robotic arm reaches a working position suitable for photographing the front axle components.

[0059] The onboard industrial control computer 7 sends motion commands to the robotic arm control box, controlling the six-axis robotic arm 9 to move the vision camera 10 sequentially to multiple preset shooting postures. Each preset posture corresponds to a key detection part on the front axle, such as the wheel, axle box, spring, shock absorber, traction motor suspension, etc.

[0060] In each shooting posture, the vision camera 10 captures high-definition images. The captured image data is transmitted to the vehicle-mounted industrial control computer 7 in real time for storage. The computer then uses a pre-set AI visual recognition algorithm to determine whether there are any abnormalities such as missing parts, damage, cracks, or oil leaks.

[0061] While the first robot is acquiring images of the front axle, the second robot, maintaining a preset distance (bogie wheelbase) from the first robot, is positioned directly below the rear axle of the same bogie.

[0062] The second robot performs the same image acquisition process as the first robot: it controls the lifting column 11 to rise to a preset height, controls the six-axis robotic arm 9 to drive the vision camera 10 to move to multiple preset shooting postures in sequence, acquires images of key parts of the rear axle, and performs storage and fault identification.

[0063] S05: When the first robot reaches the first preset endpoint, control the second robot to turn off the follow-up mode.

[0064] For example, when the first robot reaches the first preset endpoint P5, the platform system controls the second robot to turn off the follow-up mode. At this time, both the first robot and the second robot are in an independent operating state.

[0065] This technical solution uses a preset starting point and a preset distance to ensure that after the first robot completes a relative positioning, the second robot automatically completes its positioning by maintaining the preset distance, achieving simultaneous data collection from both points with a single positioning operation. Compared to traditional single-machine positioning, which requires positioning every time, this solution reduces the number of positioning operations by half, significantly improving inspection efficiency and adapting to the needs of shorter subway downtime.

[0066] In other embodiments of this application, the collaborative operation method of subway inspection robots further includes: After the first robot and the second robot reach the first preset starting point and the second preset starting point respectively, the lateral distance between the first robot and the walls on both sides of the trench is kept consistent so that the first robot and the second robot can walk along the center line of the trench.

[0067] Maintenance trenches are typically long and narrow, sunken structures, approximately 1.2 to 1.5 meters wide and 1.0 to 1.5 meters deep. They have vertical concrete or brick side walls and a hardened surface. Ramps are installed at both ends of the trench for robots to enter from the ground. The trench is narrow, allowing only one robot to pass at a time; two robots can only travel one after the other.

[0068] The walls on both sides of the trench may be equipped with facilities such as cable trays, lighting fixtures, and maintenance sockets, which will further reduce the actual passage width and increase the risk of robot collisions.

[0069] The first and second robots respectively use their onboard obstacle detection lidar 13 to detect the lateral distances to the walls on both sides of the trench.

[0070] Specifically, each robot has two obstacle detection lidars 13 arranged diagonally at the front and rear of the main body of the frame. Each obstacle detection lidar 13 has a horizontal scanning angle of 120° to 180°, which can cover the fan-shaped area in front of the robot.

[0071] As the robot travels along the trench, the obstacle detection lidar 13 continuously emits laser beams towards the walls on both sides of the trench. By measuring the time difference between laser emission and reception, the distance values ​​of the lidar reaching each point on the wall are calculated. The onboard industrial control computer 7 collects these distance data, removes noise points through a filtering algorithm, and extracts the distance values ​​of the robot's left and right walls.

[0072] The on-board industrial control computer 7 compares the detected distance values ​​of the left and right walls and calculates the offset and offset direction.

[0073] When the distance on the left is greater than the distance on the right, it indicates that the robot is veering to the right and there is a risk of colliding with the right wall. At this time, the on-board industrial control computer 7 sends a correction command to the four-wheel drive chassis, controlling the steering motors of the four sets of four-wheel drive wheel systems 8 to produce a small lateral deflection angle in the same direction, so that the robot moves laterally to the left while moving forward, until the difference between the distance on the left and the distance on the right is less than a preset threshold (such as ±2 cm).

[0074] When the distance on the left is less than the distance on the right, it indicates that the robot is veering to the left. Control the robot to move laterally to the right to correct the deviation.

[0075] During the correction process, the robot maintains its forward direction and adjusts its lateral position solely through the lateral movement capability of its four-wheel drive eight-wheel system. The omnidirectional movement characteristics of the four-wheel drive chassis allow this lateral correction to be completed smoothly without interrupting driving or changing the robot's frontal orientation.

[0076] The technical solution of this application uses a barrier-detecting lidar 13 to detect lateral distance in real time and perform dynamic correction. The two robots always maintain their centerline while traveling in the narrow trench, effectively avoiding the risk of collision with the side walls and ensuring the safety of the equipment and robots. At the same time, the consistency of the lateral positions of the two robots ensures the accuracy of laser ranging, making the distance measurement in follow-up mode more precise and reliable.

[0077] In other embodiments of this application, the collaborative operation method of subway inspection robots further includes: After the second robot turns off the follow-up mode, it controls the first and second robots to drive out of the pit and travel along both sides of the vehicle to be repaired toward the preset charging point. During the journey to the preset charging point, the first robot and the second robot are controlled to collect images from the visible components on one side of each robot.

[0078] Specifically, the onboard industrial control computer 7 of the first robot and the second robot controls the height adjustment of their respective lifting columns and controls their respective six-axis robotic arms 9 to drive the vision camera 10 to adjust to the preset shooting angle when the first robot and the second robot travel to each detection point according to the preset detection points on the vehicle side, so as to collect image information of key components on the vehicle side (such as doors, windows, vehicle body surface, headlights, etc.).

[0079] The acquired images are stored and fault identified in real time in their respective vehicle-mounted industrial control computers 7, and the identification results are transmitted back to the remote server via a wireless router.

[0080] This application's technical solution naturally completes vehicle-side inspection by utilizing the necessary path of returning to the charging point, eliminating the need for additional time and path planning. Two robots work in parallel on both sides of the vehicle, equivalent to two inspectors working simultaneously, doubling the inspection efficiency. This working mode transforms the idle time during the return trip to the charging point into effective inspection time, fully leveraging the efficiency of dual-robot collaborative operation and achieving the integrated operation goal of completing full vehicle inspection of the undercarriage and sides in a single deployment.

[0081] In a specific exemplary embodiment, after the first robot and the second robot drive out of the trench, they plan a driving path to their respective preset charging points based on their current positions and their respective preset charging points.

[0082] The location of the preset charging point is known. After the first and second robots drive out of the ramps at opposite ends of the trench, they each use their onboard multi-line positioning LiDAR5 and SLAM navigation system to obtain their precise location coordinates in the current operational environment in real time, and report the location information to the platform system via a wireless router.

[0083] Meanwhile, the platform system pre-stores an environment map of the application library, which includes key location information such as the coordinates of the preset charging points of the two robots, the location of the trench exit, and the vehicle parking area.

[0084] Based on the current location of the first robot and the coordinates of its preset charging point, the platform system uses an environmental map and corresponding algorithms to plan an optimal return path. This path prioritizes the shortest route while avoiding congested areas such as trench exits and vehicle ends.

[0085] Similarly, the platform system plans another return path based on the current position of the second robot and the coordinates of its preset charging point.

[0086] During path planning, the platform system performs intersection detection on the two paths. When an intersection or collinear overlap is detected between the two paths, the platform system adjusts the driving order of one path or adds avoidance nodes to prevent path conflicts between the two robots during their travel.

[0087] The technical solution of this application uses dynamic path planning after exiting the ditch. The platform system can intelligently plan the optimal return path based on the real-time position of the two robots and the preset charging point, so as to avoid path intersection and mutual interference between the two vehicles during the return process.

[0088] In a specific illustrative embodiment, according to the travel path, the first robot and the second robot travel along one side of the vehicle to be repaired towards their respective preset charging points in opposite directions.

[0089] Specifically, if the first robot exits from the first end of the trench, its path to the charging point is planned to be along the left side of the vehicle towards the charging point at the first end; if the second robot exits from the second end of the trench, its path to the charging point is planned to be along the right side of the vehicle towards the charging point at the second end. If the charging points are on the same side, the two robots will travel around the ends of the vehicle to ensure that they remain on opposite sides of the vehicle and in opposite directions during the return journey.

[0090] This technical solution avoids queuing at narrow exits by having two robots exit from opposite ends of the trench. When planning the path to the charging point, the two robots travel along opposite sides of the vehicle, achieving full-coverage detection on the vehicle side while avoiding path crossings and interference that could occur if the two robots travel on the same side. This control strategy of separating robots upon exiting the trench fully utilizes the resources of both robots, converting idle time during the return journey into effective detection time, achieving the effect of multiple tasks in a single deployment.

[0091] In other embodiments of this application, during the journey to a preset charging point, the first robot and the second robot are controlled to collect infrared thermal imaging data of the environment through their respective infrared cameras. The infrared thermal imaging data is analyzed in real time. When the temperature of any area exceeds a preset threshold, it is determined to be a temperature abnormality event. Based on the temperature abnormality event, the system triggers an alarm prompt.

[0092] For example, the first robot and the second robot monitor the environment using their respective infrared cameras 12 and visible light pan-tilt cameras. The visible light pan-tilt cameras perform horizontal ±360° and vertical ±90° rotation scanning along a preset trajectory, recording video, while the infrared cameras 12 acquire infrared thermal imaging data. The onboard industrial control computers 7 of the first and second robots analyze the thermal imaging data in real time. When the temperature in any area exceeds a preset threshold, it is determined to be a temperature anomaly event. The system then transmits the anomaly information back to the remote server via a wireless router and triggers an alarm.

[0093] The warehouse environment is complex, and electrical equipment and charging stations pose a risk of overheating. Traditional monitoring relies on fixed cameras, which have blind spots. This application's technical solution utilizes the mobility of an inspection robot to perform panoramic scanning and monitoring of the warehouse environment while simultaneously conducting side-mounted inspections. Infrared thermal imaging can penetrate smoke and is unaffected by light, enabling timely detection of safety hazards such as equipment overheating. Visible light recording provides visual evidence for post-incident traceability. This multi-tasking concurrent working mode fully leverages the robot's mobile perception capabilities, achieving comprehensive safety assurance through inspection as monitoring.

[0094] A second aspect of this disclosure provides a collaborative operation system for subway inspection robots, used to implement the collaborative operation method for subway inspection robots disclosed in the first aspect of this disclosure, such as... Figure 5 As shown, the subway inspection robot collaborative operation system includes platform system 1, first robot 2, and second robot 3.

[0095] Platform system 1 is used to issue control commands according to the inspection task, and set the first preset starting point P3, the second preset starting point P2 and the first preset ending point P5 according to the car body of the vehicle to be inspected. The distance between the two preset starting points P3 and P2 in the direction of the trench is the preset distance. Structural references for Robot 2 and Robot 3 Figure 5 , Figure 6 , Figure 7 Both the first robot 2 and the second robot 3 are equipped with top cameras 4 on their upper surfaces to capture key feature points of the underside of the vehicle to be inspected. The second robot 3 includes a multi-line positioning lidar 5 for real-time detection of the relative distance between it and the first robot 2. The first robot 2 and the second robot 3 communicate wirelessly.

[0096] Among them, the first robot 2 and the second robot 3 enter the pit corresponding to the vehicle to be tested according to the system control instructions, and arrive at the first preset starting point P3 and the second preset starting point P2 respectively; After both the first robot 2 and the second robot 3 reach their respective preset starting points, the first robot 2 moves to the bogie area of ​​the current carriage according to the system control command and performs relative positioning with the current carriage; according to the system control command, the second robot 3 starts the follow-up mode. In the follow-up mode, the relative distance between the second robot 3 and the first robot 2 is maintained at the preset distance.

[0097] After the first robot completes its relative positioning with the current carriage, the first robot and the second robot respectively acquire images of the undercarriage components of the current carriage according to the system control instructions. When the first robot reaches the first preset endpoint P5, the second robot ends the follow-up mode according to the system control command.

[0098] A third aspect of this disclosure provides a computer program product comprising computer program code that, when executed on a computer, causes the computer to perform the subway inspection robot collaborative operation method described in the above embodiments. This computer program can be integrated into a subway inspection robot collaborative operation system, such as a control unit of the system, for controlling a first robot and a second robot.

[0099] A fourth aspect of this disclosure provides a computer-readable storage medium storing program code that is executed by one or more processors. When the program code runs on the processors, it causes a device including one or more processors to perform the subway inspection robot collaborative operation method described in the above embodiments. The processor running this computer-readable medium can be mounted in a subway inspection robot collaborative operation system for controlling a first robot and a second robot.

[0100] It should be understood that when the modules or units described herein are implemented using software, they can be implemented in whole or in part as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0101] A fifth aspect of this disclosure provides a chip system comprising a processor, or the chip system comprising a memory and a processor, for calling computer programs or computer instructions stored in the memory to cause the processor to execute the interactive methods involved in the above embodiments. The chip system can be a single chip or a chip module composed of multiple chips. This chip system can be mounted in a subway inspection robot collaborative operation system for controlling a first robot and a second robot. Optionally, the memory of the control unit is an in-chip storage unit, such as a register or cache, or the storage unit can be an external storage unit located within a wireless access device, such as a read-only memory (ROM). ROM (Read-Only Memory) or other types of static storage devices that can store static information and instructions, such as random access memory (RAM).

[0102] Those skilled in the art will recognize that the modules, units, and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0103] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A collaborative operation method for subway inspection robots, characterized in that, include: Based on the bogie position of the vehicle to be inspected, a first preset starting point, a second preset starting point, and a first preset ending point are set, and the distance between the two preset starting points in the trench direction is the preset distance. According to the inspection task, the first robot and the second robot are controlled to enter the pit corresponding to the vehicle to be inspected and reach the first preset starting point and the second preset starting point respectively. After reaching the preset starting point, the first robot is controlled to move to the bogie area of ​​the current carriage and perform relative positioning with the current carriage; the second robot is controlled to start the follow-up mode, in which the relative distance between the second robot and the first robot is maintained at the preset distance. After the first robot completes its relative positioning with the current carriage, it controls the first robot and the second robot to respectively acquire images of the undercarriage components of the current carriage; When the first robot reaches the first preset endpoint, the second robot is controlled to turn off the follow-up mode.

2. The collaborative operation method for subway inspection robots according to claim 1, characterized in that, The method further includes: After the first robot and the second robot reach the first preset starting point and the second preset starting point respectively, the lateral distance between the first robot and the second robot and the walls on both sides of the trench is kept consistent so that the first robot and the second robot walk along the center line of the trench.

3. The method for collaborative operation of subway inspection robots according to claim 1, characterized in that, The preset distance is the wheelbase of the bogie of the vehicle to be inspected.

4. The collaborative operation method for subway inspection robots according to claim 1, characterized in that, Also includes: After the second robot closes the follow-up mode, it controls the first robot and the second robot to drive out of the ditch and drive along both sides of the vehicle to be repaired toward the preset charging point. During the journey to the preset charging point, the first robot and the second robot are controlled to capture images of the visible components on one side of each robot.

5. The method for collaborative operation of subway inspection robots according to claim 4, characterized in that, After the first robot and the second robot leave the trench, they plan their driving routes to their respective preset charging points based on their current positions and preset charging points.

6. The method for collaborative operation of subway inspection robots according to claim 4, characterized in that, According to the travel path, the first robot and the second robot travel along one side of the vehicle to be repaired towards their respective preset charging points in opposite directions.

7. The method for collaborative operation of subway inspection robots according to claim 4, characterized in that, During the journey to the preset charging point, the first and second robots are controlled to collect infrared thermal imaging data of the environment through their respective infrared cameras. The infrared thermal imaging data is analyzed in real time. When the temperature of any area exceeds a preset threshold, it is determined to be a temperature abnormality event. Based on the temperature abnormality event, the system triggers an alarm prompt.

8. The method for collaborative operation of subway inspection robots according to claim 1, characterized in that, Controlling the second robot to activate the follow-up mode, wherein in the follow-up mode, the relative distance between the second robot and the first robot is maintained at the preset distance, including: The second robot detects its distance from the first robot using a multi-line positioning lidar. Based on its distance from the first robot, the second robot calculates its compensation value relative to the first robot in the trench direction; Based on the compensation value, the second robot accelerates or decelerates to maintain the relative distance at the preset distance.

9. The collaborative operation method of subway inspection robots according to claim 1, characterized in that, The process of relative positioning with respect to the current carriage includes: The first robot is controlled to acquire an image of the bogie area of ​​the current carriage through the top camera. The acquired image of the bogie area of ​​the current carriage is matched and identified with a pre-stored bogie center pin template to obtain the pixel coordinates of the bogie center pin in the image. Based on the pixel coordinates and the pre-stored installation height of the top camera, the relative pose deviation between the first robot and the center pin of the bogie is calculated through coordinate transformation; Based on the positional deviation, the first robot is controlled to make micro-adjustments, so that the first robot moves directly below the pre-stored shooting point of the bogie center pin.

10. A collaborative operation system for subway inspection robots, characterized in that, The system employing the collaborative operation method for subway inspection robots according to any one of claims 1-9 includes: The platform system is configured to issue control commands according to the inspection task, and set a first preset starting point, a second preset starting point, and a first preset ending point according to the carriage of the vehicle to be inspected. The distance between the two preset starting points in the direction of the trench is a preset distance. Both the first and second robots are equipped with top-mounted cameras on their upper surfaces to capture key feature points of the underside of the vehicle to be inspected. The second robot includes a multi-line positioning LiDAR for real-time detection of the relative distance to the first robot. The first and second robots communicate wirelessly. The first and second robots are configured as follows: According to the system control instructions, the vehicle enters the pit corresponding to the vehicle to be detected and arrives at the first preset starting point and the second preset starting point respectively; Upon reaching the preset starting point, the first robot moves to the bogie area of ​​the current carriage according to the system control command and performs relative positioning with the current carriage; the second robot starts the follow-up mode, in which the relative distance between the second robot and the first robot is maintained at the preset distance; After the first robot completes its relative positioning with the current carriage, the first robot and the second robot respectively acquire images of the undercarriage components of the current carriage according to the system control instructions; When the first robot reaches the first preset endpoint, the second robot terminates the follow-up mode according to the system control command.

Images