A dynamic vehicle axle detection, inspection and positioning system

Abstract

The present application relates to the technical field of inspection of motor train unit, and particularly relates to a motor train axle detection and inspection positioning system. The system comprises a robot body, a sensing module, a walking driving mechanism, an upper computer control cabinet and a mechanical arm module. The sensing module comprises at least two 3D line laser sensors and at least two point laser distance sensors, which are respectively used for scanning the profile data of the bottom of the motor train and measuring the distance data of each feature point. The output end of the walking driving mechanism is provided with an encoder for real-time acquisition of the displacement information of the robot. The upper computer control cabinet is configured to: establish a positioning starting point by recognizing the head of the motor train through the point laser; synchronously receive the profile data and the distance data and associate the displacement information of the encoder; generate a wheel axle positioning track and a wheel axle height change track through fusion processing; and drive the robot to position and control the height compensation of the mechanical arm to perform fine scanning. The present application realizes high-precision positioning of the motor train axle through multi-sensor fusion, and effectively improves the precision and anti-interference ability of the inspection positioning.

Description

Technical Field

[0001] This invention relates to the field of high-speed train inspection technology, and in particular to a high-speed train axle detection and inspection positioning system. Background Technology

[0002] With the development of the high-speed rail industry, the demand for safety inspections of high-speed trains is increasing, and the safety requirements are becoming more stringent, especially for critical safety components. Therefore, a more efficient, accurate, and flexible detection and positioning method is needed to achieve accurate detection of critical components.

[0003] The high-definition acquisition cameras configured in the currently commonly used automated inspection equipment for the undercarriage of high-speed trains have certain limitations in terms of shooting angle and shooting method. Moreover, due to the complexity of the undercarriage structure of high-speed trains, key inspection parts require shooting from specific angles and have very high repeatability accuracy in order to obtain accurate, clear and unobstructed image data. Therefore, the positioning method has very high requirements.

[0004] Existing detection and positioning methods include mechanical ranging and laser positioning. The former offers higher positioning accuracy, but cumulative errors occur during long-distance detection and positioning, resulting in lower positioning accuracy, limited operational coverage, and lower positioning efficiency. The latter, relying solely on laser positioning, offers high efficiency, but its positioning accuracy is low and it is significantly affected by environmental interference. Summary of the Invention

[0005] This invention provides a high-speed train axle inspection and positioning system, which at least partially solves the technical problems of low positioning accuracy, poor anti-interference ability, and inability to achieve repeatable positioning in the prior art.

[0006] Embodiments of this disclosure provide a vehicle axle detection, inspection, and positioning system, comprising: The robot itself is movably mounted on the inspection track at the bottom of the train. The sensing module, installed on the robot body, includes: At least two 3D line laser sensors are configured to scan the outline data of the underside of the train vertically upwards during travel; At least two point laser distance sensors are configured to measure distance data to various feature points on the bottom of the train vertically upwards; A walking drive mechanism is configured to drive the robot body to walk along the inspection track. An encoder is installed at the output end of the walking drive mechanism and is configured to acquire the displacement information of the robot body in real time. The host computer control cabinet is communicatively connected to the 3D line laser sensor, the point laser distance sensor and the encoder respectively; At least one robotic arm module is installed on the robot body and connected to the host computer control cabinet for performing fine cleaning actions; The host computer control cabinet is configured to identify the reference features at the bottom of the vehicle front through the point laser distance sensor and record the current value of the encoder as the positioning starting point. During the movement, the contour data collected by the 3D line laser sensor and the distance data collected by the point laser distance sensor are received simultaneously, and the displacement information of the encoder is correlated in real time. The contour data and the distance data are fused together. Based on the fusion result and the displacement information, the horizontal position and vertical height of each bogie wheel axle are calculated to generate wheel axle positioning trajectory and wheel axle height change trajectory. The robot arm module is moved to the target wheel axle position by controlling the traveling drive mechanism according to the wheel axle positioning trajectory, and the robot arm module is height compensated according to the wheel axle height change trajectory to perform fine sweeping action.

[0007] The technical solution provided in this application brings at least the following beneficial effects: high-precision positioning is achieved through multi-sensor collaboration. 3D line laser provides contour data, point laser provides precise distance, and encoder provides displacement information. The fusion of these three technologies eliminates the limitations of a single sensor, enabling high-precision positioning of the train axle.

[0008] In other embodiments of this application, the step of identifying the reference features at the bottom of the vehicle front using the point laser distance sensor and recording the current encoder value as the positioning starting point includes: The height information of the front of the vehicle is obtained based on the point laser distance sensor; The height information is compared with a preset threshold. The vehicle's front position is identified based on the comparison between the height information and a preset threshold, and the current encoder value is recorded. The current encoder value is converted into the position of the robot body on the inspection track as the positioning starting point.

[0009] The technical solution provided in this application brings at least the following benefits: by comparing the height information with a preset threshold to identify the position of the vehicle head, the encoder value is converted into a positioning starting point, an absolute position reference is established, and the cumulative error in long-distance inspection is eliminated.

[0010] In other embodiments of this application, two point laser distance sensors are arranged side by side along a direction perpendicular to the forward movement of the robot body. The distance data includes characteristic data of the bogie wheel axle and interference characteristic data similar to the bogie wheel axle characteristics. One point laser distance sensor is configured to collect the characteristic data of the bogie wheel axle, and the other point laser distance sensor is configured to collect interference characteristic data similar to the bogie wheel axle characteristics.

[0011] The technical solution provided in this application brings at least the following benefits: by collecting wheel axle features and interference features in groups using point lasers, and using the interference features as negative samples for comparison and verification, the misidentification rate caused by interference components in the complex environment at the bottom of the train is reduced.

[0012] In other embodiments of this application, the step of fusing the contour data and the distance data, calculating the horizontal position and vertical height of each bogie axle based on the fusion result and the displacement information, and generating axle positioning trajectories and axle height change trajectories includes: The characteristic data of the bogie wheel axle are compared with the interference characteristic data; Based on the comparison results, the interference features corresponding to the interference feature data are eliminated, and the true feature data of the bogie wheel axle is obtained. The horizontal position and vertical height of each bogie axle are calculated based on the actual axle feature data and the displacement information.

[0013] The technical solution provided in this application brings at least the following beneficial effects: after eliminating interference through feature comparison, the position and height of the wheel axle are calculated, and the sensor data and displacement information are fused to achieve simultaneous and accurate calculation of the horizontal position and vertical height of the wheel axle.

[0014] In other embodiments of this application, the step of comparing the feature data of the bogie wheel axle with the interference feature data, eliminating the interference features corresponding to the interference feature data based on the comparison result, and obtaining the true feature data of the bogie wheel axle includes: The host computer control cabinet marks the arc features in the contour data collected by the 3D line laser sensor as candidate wheel axle positions; The host computer control cabinet performs spatial matching between the distance data collected by the point laser distance sensor and the position of the candidate wheel axle; The host computer control cabinet eliminates the interference features corresponding to the interference feature data based on the spatial matching results, and obtains the true feature data of the bogie wheel axle.

[0015] The technical solution provided in this application brings at least the following benefits: by marking candidate positions with line laser arc features and matching point laser distance data with candidate positions in space, a two-level filtering mechanism is achieved, further eliminating false features and improving positioning reliability.

[0016] In other embodiments of this application, the step of fusing the contour data and the distance data, calculating the horizontal position and vertical height of each bogie axle based on the fusion result and the displacement information, and generating axle positioning trajectories and axle height change trajectories includes: The encoder values ​​are acquired in real time; The encoder value is converted into the actual running distance of the robot body on the inspection track; Based on the fusion processing results and the actual running distance, the horizontal position and vertical height of each bogie wheel axle are calculated, and wheel axle positioning trajectory and wheel axle height change trajectory are generated.

[0017] The technical solution provided in this application brings at least the following benefits: by converting encoder values ​​into actual running distances and associating them with fusion results, the precise binding of sensor feature data and spatial location is achieved, providing a quantifiable position benchmark for trajectory generation.

[0018] In other embodiments of this application, the scanning direction of the 3D line laser sensor is perpendicular to the forward direction of the robot body, the measurement direction of the point laser distance sensor is parallel to the scanning direction of the 3D line laser sensor, and the measurement point of the point laser distance sensor is located on or near the scanning line of the 3D line laser sensor.

[0019] The technical solution provided in this application brings at least the following beneficial effects: by unifying the spatial layout of sensors (vertically upward, parallel installation, and measurement points located on the scanning line), it ensures that data from multiple sensors share the same spatial reference system, thus eliminating coordinate transformation errors.

[0020] In other embodiments of this application, the walking drive mechanism includes a servo motor, a reducer, and a drive output shaft. The output end of the servo motor is connected to the input end of the reducer, and the output end of the reducer is connected to the drive output shaft. The drive output shaft is connected to the robot body and is configured to drive the robot body to walk. The encoder is fixed to the output end of the reducer and is configured to detect the rotation angle and number of rotations of the drive output shaft to obtain the displacement information.

[0021] The technical solution provided in this application brings at least the following benefits: by using a three-stage transmission structure and installing an encoder at the output end of the reducer, the actual rotation of the drive wheel can be directly measured, eliminating errors in the intermediate links of the transmission chain and ensuring the authenticity of displacement measurement.

[0022] In other embodiments of this application, the robotic arm module includes a lifting module, a robotic arm, and a precision scanning camera. The lifting module is vertically mounted on the robot body, the robotic arm is mounted on the movable end of the lifting module, and the precision scanning camera is mounted on the end effector of the robotic arm. The lifting module is configured to drive the robotic arm to perform height interpolation in the vertical direction according to the trajectory of the wheel axle height change, so that the precision scanning camera maintains a constant working distance from the target wheel axle. The robotic arm module is configured such that the direction of movement of the robot body during precision scanning inspection is opposite to the direction of movement during line scanning inspection. The technical solution provided in this application brings at least the following benefits: by actively compensating for the trajectory of height change through the lifting module, the fine scanning camera and the wheel axle maintain a constant working distance, and reverse fine scanning avoids interference, thus ensuring the clarity of the fine scanning image and the safety of the operation.

[0023] In other embodiments of this application, the host computer control cabinet is configured to establish a global coordinate system with the positioning starting point as the origin, generate an absolute position trajectory of the wheel axle positioning trajectory and the wheel axle height change trajectory with reference to the global coordinate system, and associate and store the absolute position trajectory with the train vehicle identification number.

[0024] The technical solution provided in this application brings at least the following benefits: by establishing a global coordinate system to generate an absolute position trajectory and storing it in association with the vehicle identification number, it enables repeated maintenance after a single positioning, and the trajectory can be directly called to perform fine scanning when the same vehicle re-enters the warehouse.

[0025] Compared with the prior art, the present invention has the following beneficial effects: 1. By deeply integrating the three technologies—3D line laser sensor to acquire vehicle undercarriage contour data, point laser distance sensor to acquire precise distances to key points, and encoder to acquire displacement information—the system overcomes the shortcomings of single sensors in the complex environment under the train, such as low positioning accuracy and poor anti-interference capability. The positioning accuracy is significantly higher than that of pure laser positioning solutions.

[0026] 2. By grouping multiple point laser distance sensors to collect wheel axle features and interference features and comparing and eliminating them, the false identification rate caused by interference components in the complex environment under the train is greatly reduced.

[0027] 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

[0028] 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.

[0029] Figure 1 This is a schematic diagram of the structure of the train axle detection, inspection and positioning system according to an embodiment of this application; Figure 2 This is the flow chart of the control method for the train axle detection, inspection and positioning system according to an embodiment of this application. Figure 1 ; Figure 3 This is the flow chart of the control method for the train axle detection, inspection and positioning system according to an embodiment of this application. Figure 2 ; Figure 4 This is the flow chart of the control method for the train axle detection, inspection and positioning system according to an embodiment of this application. Figure 3 ; Figure 5 This is a schematic diagram of the running drive structure of the train axle detection, inspection and positioning system according to an embodiment of this application; In the above figures, 1. Robot body; 2. First 3D line laser sensor; 3. Second 3D line laser sensor; 4. First point laser distance sensor; 5. Second point laser distance sensor; 6. Third point laser distance sensor; 7. Fourth point laser distance sensor; 8. Host computer control cabinet; 9. First robotic arm module; 10. Second robotic arm module; 11. Third robotic arm module; 12. First line scanning camera; 13. Second line scanning camera; 14. Walking drive mechanism; 15. Servo motor; 16. Reducer; 17. Drive output shaft; 18. Encoder. Detailed Implementation

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] In the field of high-speed train axle inspection, existing technologies mainly employ mechanical ranging or single laser sensor positioning to locate the inspection device on the underside of the train. Mechanical ranging calculates the travel distance by measuring the number of wheel rotations, offering high positioning accuracy, but it suffers from cumulative errors during long-distance detection and positioning, and its operational range is limited. While single laser sensor positioning offers high efficiency, its accuracy is low and it is susceptible to interference from the complex structure of the train's underside and ambient light. As the requirements for accuracy in high-speed train underside inspection continue to increase, the above methods are insufficient to accurately identify axle positions, eliminate cumulative errors, and resist environmental interference. Furthermore, the underside of the train contains various interfering components with similar axle contours (such as shock absorber mounting brackets, lateral stop brackets, and brake hangers). Existing positioning methods cannot effectively distinguish between the axle and these interfering features, leading to positioning trajectory deviations and affecting the accuracy of subsequent fine-sweeping operations.

[0036] To address the aforementioned technical problems, this application proposes a high-speed train axle detection, inspection, and positioning system. The specific implementation methods of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. like Figure 1 As shown, the present invention provides a high-speed train axle detection, inspection and positioning system, including a robot body 1, a sensing module, a traveling drive mechanism 14, a host computer control cabinet 8 and at least one robotic arm module.

[0037] The robot body 1 is movably mounted on the inspection track at the bottom of the train. It has a rectangular frame structure and drive wheels at the bottom, enabling it to move along the track direction on the inspection track at the bottom of the train.

[0038] The sensing module, mounted on the robot body 1, includes at least two 3D line laser sensors and at least four point laser distance sensors. The 3D line laser sensors, mounted on the robot body 1, scan vertically upwards during movement, emitting line laser beams towards the underside of the train and receiving reflected light signals to acquire cross-sectional contour data of various components on the underside of the train along the scanning lines. Each 3D line laser sensor acquires one cross-sectional contour line per scan. As the robot body 1 moves forward, it continuously scans to acquire a series of cross-sectional contour lines, which are then stitched together to form the three-dimensional contour data of the train's underside.

[0039] At least two point laser distance sensors are mounted on the top of the robot body 1, with the measurement direction vertically upward. Each point laser distance sensor independently emits a single laser point and receives the reflected light, measuring the precise distance value from the sensor to the laser illumination point, and outputting a single distance value. The two point laser distance sensors work simultaneously, allowing precise distance values ​​to be obtained at multiple different locations at the same time.

[0040] The walking drive mechanism 14 is installed at the bottom of the robot body 1 and is used to drive the robot body 1 to walk along the inspection track. An encoder 18 is set at the output end of the walking drive mechanism 14, which is used to acquire the displacement information of the robot body 1 in real time.

[0041] The host computer control cabinet 8 is installed on the robot body 1 and is communicatively connected to the 3D line laser sensor, point laser distance sensor, and encoder 18. The host computer control cabinet 8 includes an industrial computer and a PLC controller.

[0042] At least one robotic arm module is installed on the robot body 1 and connected to the host computer control cabinet 8 for performing fine sweeping actions.

[0043] like Figure 2 As shown, the positioning control process executed by the host computer control cabinet 8 is as follows: S01: Identify the reference features at the bottom of the vehicle front using a point laser distance sensor and record the current encoder 18 value as the positioning starting point.

[0044] S02: During the movement, the host computer control cabinet 8 synchronously receives the contour data collected by the 3D line laser sensor and the distance data collected by the point laser distance sensor, and associates it with the displacement information of the encoder 18 in real time.

[0045] S03: The host computer control cabinet 8 fuses the contour data and distance data, calculates the horizontal position and vertical height of each bogie wheel axle based on the fusion processing result and displacement information, and generates wheel axle positioning trajectory and wheel axle height change trajectory.

[0046] S04: The host computer control cabinet 8 controls the traveling drive mechanism 14 to move the robotic arm module to the target axle position according to the wheel axle positioning trajectory, and controls the robotic arm module to perform height compensation according to the wheel axle height change trajectory to perform fine sweeping action.

[0047] This technical solution achieves high-precision positioning of train axles through the collaborative operation of a 3D line laser sensor, a point laser distance sensor, and an encoder. The 3D line laser sensor provides continuous cross-sectional contour data, capturing the arc shape features of the axle; the point laser distance sensor provides precise distance values ​​for key points, verifying and supplementing the line laser data; and the encoder provides precise displacement information, binding each frame of sensor data to its spatial position. The data from the three sensors are fused in a host computer, overcoming the shortcomings of single sensors in the complex environment under the train, such as low positioning accuracy and poor anti-interference capability. Through multi-sensor fusion, this technical solution can accurately identify the axle position in environments with various interfering components under the train, while simultaneously acquiring the horizontal position and vertical height of the axle, providing a precise positioning reference for subsequent fine scanning. Compared to pure laser positioning schemes, this invention offers higher positioning accuracy and stronger resistance to environmental interference; compared to pure mechanical ranging schemes, this implementation has no cumulative error and higher operational efficiency.

[0048] In other embodiments of this application, such as Figure 3 As shown, a point laser distance sensor identifies the reference features at the bottom of the vehicle's front end and records the current encoder value as the positioning starting point, including: S11: Obtain the height information of the front of the vehicle based on the point laser distance sensor; Specifically, such as Figure 1 As shown, the robot body 1 moves along the inspection track, from the outside of the train to the bottom of the front. A point laser distance sensor emits a laser beam vertically upwards and receives the reflected light, measuring the distance between the sensor and the object above it in real time. The host computer control cabinet 8 collects the distance values ​​from the point laser distance sensor at a fixed frequency.

[0049] S12: Compare the height information with a preset threshold; When robot body 1 is located outside the train, there are no obstructions above the point laser distance sensor. At this time, the measured distance value is either too large or invalid. The host computer control cabinet 8 compares this distance value with a preset threshold. If the current distance value is greater than the preset threshold, it indicates that robot body 1 has not yet entered the bottom of the train's front.

[0050] As robot body 1 continues to move forward, and its top enters below the bottom of the vehicle's front, the distance value measured by the point laser distance sensor suddenly decreases. The host computer control cabinet 8 detects that the distance value has changed from greater than a preset threshold to less than a preset threshold, and determines that robot body 1 has now entered the bottom of the vehicle's front.

[0051] S13: Identify the position of the vehicle head based on the comparison between the height information and the preset threshold, and record the current encoder value; When the host computer control cabinet 8 determines that the vehicle has entered the bottom of the front of the vehicle, it immediately reads the current value of the encoder 18.

[0052] S14: Convert the current encoder value into the position of the robot body 1 on the inspection track as the positioning starting point.

[0053] The host computer control cabinet 8 converts the current position of the robot body 1 into its position on the inspection track based on the current value of the encoder 18, defines the position as the positioning starting point, and uses the encoder value at this position as the reference zero point for all subsequent position calculations.

[0054] This application's technical solution achieves automatic identification of the train's front position by comparing the real-time height information measured by a point laser distance sensor with a preset threshold. The point laser distance sensor is unaffected by changes in lighting and provides a continuous distance value change curve, ensuring reliable identification. By setting a preset threshold, interference from factors such as ground undulations and sensor installation errors can be eliminated. Using the encoder value at the moment the train's front is identified as the positioning starting point, an absolute coordinate reference strongly correlated with the train's physical position is established. All subsequent wheel and axle positions are calculated relative to this starting point, eliminating accumulated errors caused by wheel diameter errors, slippage, and other factors during long-distance inspections, ensuring consistent positioning throughout the entire train.

[0055] In other embodiments of this application, two point laser distance sensors are arranged side by side along a direction perpendicular to the forward movement of the robot body. The distance data includes characteristic data of the bogie wheel axle and interference characteristic data similar to the bogie wheel axle characteristics. One point laser distance sensor is used to collect the characteristic data of the bogie wheel axle, and the other point laser distance sensor is used to collect interference characteristic data similar to the bogie wheel axle characteristics.

[0056] Specifically, such as Figure 1 As shown, the first laser distance sensor 4, the second laser distance sensor 5, the third laser distance sensor 6, and the fourth laser distance sensor 7 are arranged side by side on the top of the robot body 1 along a direction perpendicular to the forward direction of the robot body 1. The spacing between each sensor is preset according to the axial width of the wheel axle of the train bogie.

[0057] The first laser distance sensor 4 and the second laser distance sensor 5 are used to collect characteristic data of the bogie wheel axle; the third laser distance sensor 6 and the fourth laser distance sensor 7 are used to collect interference characteristic data that are similar to the characteristics of the bogie wheel axle.

[0058] When the four point laser distance sensors collect distance data vertically upwards, their different installation positions result in each sensor illuminating different lateral positions on the bottom of the vehicle. As the robot body 1 moves, the four point laser distance sensors synchronously and continuously sample, each outputting a curve showing the change in distance value over time.

[0059] The bogie axles at the bottom of the train are cylindrical with a circular cross-section. When the robot body 1 moves from one side of the axle to the other, the distance values ​​measured by the point laser distance sensor directly opposite the axle exhibit a pattern of first decreasing and then increasing. The first point laser distance sensor 4 and the second point laser distance sensor 5 are installed directly opposite the axle, so the sequence of distance values ​​they collect exhibits the aforementioned arc-shaped change pattern, including the arc-shaped edge features of the axle.

[0060] Besides the wheel axles, the underside of the train also contains components such as shock absorber mounting seats, lateral stop seats, and brake hangers. Some of these components also have rounded edges or protruding structures. The third-point laser distance sensor 6 and the fourth-point laser distance sensor 7 are installed on both sides of the wheel axles, with their illumination points located outside the wheel axles. Therefore, when the robot body 1 passes the wheel axle position, the distance value sequence collected by the third-point laser distance sensor 6 and the fourth-point laser distance sensor 7 will not exhibit the changing pattern of the wheel axle's rounded edge. These data reflect the characteristics of interference components on the train's underside that are similar to the wheel axle features, and are therefore used to collect interference characteristic data similar to the bogie wheel axle features.

[0061] This technical solution achieves simultaneous measurement of different lateral positions on the underside of the train by arranging multiple point laser distance sensors side-by-side perpendicular to the direction of travel and dividing them into two groups to collect wheel axle features and interference features respectively. The underside of the train contains various interference components with contour features similar to those of the wheel axle; a single sensor, relying solely on contour scanning, can easily misidentify these interference components as wheel axles. By acquiring the distance change curve (exhibiting an arc pattern) directly above the wheel axle using the first and second point laser sensors, and simultaneously acquiring the distance change curves at interference components on both sides of the wheel axle using the third and fourth point laser sensors, the host computer can compare the two sets of data. This dual-channel acquisition mechanism of "target features + interference features" provides reliable negative sample data for subsequent interference elimination, significantly reducing the false identification rate.

[0062] In other embodiments of this application, such as Figure 4 As shown, the contour data and distance data are fused. Based on the fusion results and displacement information, the horizontal position and vertical height of each bogie axle are calculated, generating axle positioning trajectories and axle height change trajectories, including: S31: Compare the characteristic data of the bogie wheel axle with the interference characteristic data; S32: Based on the comparison results, eliminate the interfering features corresponding to the interfering feature data to obtain the true feature data of the bogie wheel axle; In a specific illustrative embodiment, the characteristic data of the bogie wheel axle is compared with the interfering characteristic data. Based on the comparison result, the interfering characteristics corresponding to the interfering characteristic data are eliminated to obtain the true characteristic data of the bogie wheel axle, including: The host computer control cabinet marks the arc features in the contour data collected by the 3D line laser sensor as candidate wheel axle positions; The host computer control cabinet performs spatial matching between the distance data collected by the point laser distance sensor and the candidate wheel axle position; The host computer control cabinet eliminates the interference features corresponding to the interference feature data based on the spatial matching results, and obtains the true feature data of the bogie wheel axle.

[0063] Specifically, such as Figure 1 As shown, the host computer control cabinet 8 receives contour data from the 3D line laser sensor, distance data from four point laser distance sensors, and displacement information from the encoder 18. The host computer control cabinet 8 divides the distance data from the point laser distance sensors into two groups: the data collected by the first point laser distance sensor 4 and the second point laser distance sensor 5 are classified as bogie wheel axle characteristic data, and the data collected by the third point laser distance sensor 6 and the fourth point laser distance sensor 7 are classified as interference characteristic data.

[0064] The host computer control cabinet 8 compares the characteristic data of the bogie wheel axle with the interference characteristic data.

[0065] The host computer control cabinet 8 extracts arc features from the contour data collected by the 3D line laser sensor and marks them as candidate wheel axle positions. When the first laser distance sensor 4 and the second laser distance sensor 5 show a change pattern of the distance value corresponding to the wheel axle arc at the candidate position, first decreasing and then increasing, while the third laser distance sensor 6 and the fourth laser distance sensor 7 do not show the same change pattern at the same position, the host computer control cabinet 8 determines that the candidate position is the true wheel axle position.

[0066] This application's technical solution employs a two-stage filtering mechanism—marking candidate positions with arc features and spatially matching point laser distance data—to achieve precise verification of candidate wheel axle positions. The first stage utilizes contour data from a 3D line laser sensor for rapid generalization, marking all possible arc features as candidates to ensure no omissions. The second stage uses point laser distance data for precise verification of candidate positions, eliminating interfering components through waveform similarity comparison to ensure no false detections. This two-stage architecture of "coarse identification + fine verification" significantly improves positioning accuracy while maintaining detection recall. Simultaneously, spatial matching compares point laser distance data and 3D line laser contour data within the same spatial reference frame, eliminating coordinate transformation errors and ensuring the reliability of the verification results.

[0067] S33: Calculate the horizontal position and vertical height of each bogie axle based on the actual axle characteristic data and displacement information.

[0068] After eliminating interfering features based on the comparison results, the host computer control cabinet 8 calculates the horizontal position and vertical height of the bogie axle based on the verified true axle feature data. The host computer control cabinet 8 takes multiple distance values ​​from the first laser distance sensor 4 and the second laser distance sensor 5 in the true axle feature data and uses a circular arc fitting algorithm to fit the center coordinates of the arc. The horizontal value of this center coordinate is the horizontal position of the axle, and the vertical value is the vertical height of the axle.

[0069] This application's technical solution compares the feature data of the bogie wheel axle with interference feature data. Based on the preliminary identification of candidate wheel axles using 3D line laser contour data, it verifies and filters the candidate results using two sets of point laser distance data. When the first and second laser points show the wheel axle's arc pattern while the third and fourth laser points do not, the candidate is confirmed as the true feature of the wheel axle, eliminating misidentification caused by interfering components at the bottom of the train. After confirming the true wheel axle feature, an arc fitting algorithm is used to accurately calculate the wheel axle's center position and height, achieving simultaneous and accurate calculation of the wheel axle's horizontal position and vertical height. This process integrates "feature comparison" and "position calculation" into the same fusion processing flow, ensuring the accuracy and reliability of the positioning results.

[0070] In other embodiments of this application, contour data and distance data are fused, and the horizontal position and vertical height of each bogie axle are calculated based on the fusion processing result and displacement information to generate axle positioning trajectories and axle height change trajectories, including: Obtain encoder values ​​in real time; The encoder values ​​are converted into the actual running distance of the robot body on the inspection track; Based on the fusion processing results and the actual running distance, the horizontal position and vertical height of each bogie wheel axle are calculated, and wheel axle positioning trajectory and wheel axle height change trajectory are generated.

[0071] Specifically, during the movement of the robot body 1, the host computer control cabinet 8 acquires the values ​​of the encoder 18 in real time. Each rotation of the encoder 18 outputs a fixed number of pulse signals. The host computer control cabinet 8 receives the pulse signals output by the encoder 18 in real time and counts the pulses.

[0072] The host computer control cabinet 8 stores relevant parameters of the robot body 1's walking mechanism in advance. When the host computer control cabinet 8 obtains the current value of the encoder 18, it converts the value into the actual running distance of the robot body 1 on the inspection track according to the preset conversion relationship. The specific conversion method is as follows: the host computer control cabinet 8 calculates the distance that the drive wheel rolls on the inspection track based on the pulse count value of the encoder 18, the number of pulses per revolution of the encoder 18, and the transmission ratio parameters of the robot body 1's walking mechanism. This distance is the actual running distance of the robot body 1.

[0073] During each data fusion process, the host computer control cabinet 8 synchronously acquires the current encoder 18 value and converts it into the actual running distance. This actual running distance is used as the horizontal position label of the current frame data and is stored in association with the wheel axle feature data obtained from the fusion process. When the fusion process confirms the existence of a wheel axle at a certain position, the host computer control cabinet 8 records the actual running distance corresponding to that position as the horizontal position coordinate of the wheel axle.

[0074] Throughout the entire line scanning process, the host computer control cabinet 8 continuously performs the aforementioned conversion and association operations, forming a sequence of horizontal coordinates for multiple wheel axle positions. Combining the vertical heights of each wheel axle obtained through fusion processing, the host computer control cabinet 8 generates wheel axle positioning trajectories (including the horizontal positions of each wheel axle) and wheel axle height change trajectories (including the vertical heights of each wheel axle).

[0075] This application's technical solution achieves precise binding between sensor feature data and spatial position by converting the encoder's pulse count value into the robot's actual running distance on the inspection track. A definite conversion relationship exists between the encoder output and the robot's walking distance. This conversion transforms the encoder's raw pulse values ​​into distance values ​​with clear physical meaning, allowing the host computer to describe the wheel axle position in meters or millimeters. After associating and storing the actual running distance with the fusion processing results, each wheel axle feature corresponds to a precise spatial coordinate, providing a reliable data foundation for subsequently generating repeatable positioning trajectories.

[0076] In other embodiments of this application, the scanning direction of the 3D line laser sensor is perpendicular to the forward direction of the robot body, the measurement direction of the point laser distance sensor is parallel to the scanning direction of the 3D line laser sensor, and the measurement point of the point laser distance sensor is located on or near the scanning line of the 3D line laser sensor.

[0077] like Figure 1 As shown, a 3D line laser sensor is mounted on the robot body 1, and its scanning direction is perpendicular to the forward direction of the robot body 1 (i.e., along the lateral direction of the inspection track). Specifically, the 3D line laser sensor emits a line laser beam vertically upwards towards the bottom of the vehicle, and the extension direction of the line laser beam is the lateral direction of the inspection track (perpendicular to the forward direction of the robot body 1), forming a laser line plane perpendicular to the forward direction. When the robot body 1 moves along the forward direction, this laser line plane sequentially scans each cross-section of the bottom of the vehicle, obtaining a series of cross-sectional contour data.

[0078] A point laser distance sensor is mounted on the robot body 1, and its measurement direction is parallel to the scanning direction of the 3D line laser sensor. Specifically, the point laser distance sensor emits a point laser beam vertically upward, and its measurement direction (vertically upward) is parallel to the scanning direction (vertically upward) of the 3D line laser sensor.

[0079] Meanwhile, the measurement point of the point laser distance sensor is located on or near the scanning line of the 3D line laser sensor. Since the scanning line of the 3D line laser sensor is a straight line extending laterally along the inspection track, the laser point emitted vertically upwards by the point laser distance sensor illuminates the bottom of the train precisely on or very close to this lateral scanning line. This spatial arrangement ensures that the distance data measured by the point laser distance sensor and the contour data of the 3D line laser sensor at that location correspond to the same lateral position on the bottom of the train. When performing data fusion processing, the host computer control cabinet 8 can directly spatially align the distance value collected by the point laser distance sensor with the contour data of the 3D line laser sensor at that lateral position, without requiring additional coordinate transformation.

[0080] This application's technical solution unifies the scanning direction of the 3D line laser sensor and the measurement direction of the point laser distance sensor to be vertically upward and parallel, and places the measurement point of the point laser on the scanning line of the line laser, allowing all sensor data to share the same spatial reference frame. The 3D line laser sensor acquires cross-sectional contour data, while the point laser distance sensor acquires precise single-point distance values ​​at the same lateral position, naturally aligning the two in space. This spatial layout eliminates coordinate transformation errors between different sensor data, enabling the host computer to directly match and fuse the point laser distance values ​​with the line laser contour data point by point. Simultaneously, the vertically upward measurement direction ensures the shortest laser beam path and minimal obstruction, effectively avoiding measurement errors caused by oblique projection.

[0081] In other embodiments of this application, the walking drive mechanism includes a servo motor, a reducer, and a drive output shaft. The output end of the servo motor is connected to the input end of the reducer, the output end of the reducer is connected to the drive output shaft, and the drive output shaft is connected to the robot body for driving the robot body to walk. An encoder is fixed to the output end of the reducer for detecting the rotation angle and number of rotations of the drive output shaft to obtain displacement information.

[0082] like Figure 5 As shown, the traveling drive mechanism 14 includes a servo motor 15, a reducer 16, and a drive output shaft 17.

[0083] Servo motor 15 is mounted on the bottom frame of robot body 1, with its output end extending out of the servo motor 15 housing. The output end of servo motor 15 is connected to the input end of reducer 16. The output end of servo motor 15 outputs high-speed rotational motion, which is input to reducer 16.

[0084] The reducer 16 is mounted on one side of the output end of the servo motor 15, and its input end is connected to the output end of the servo motor 15. The reducer 16 has an internal gear reduction mechanism that converts the input high-speed rotary motion into low-speed, high-torque rotary motion, which is then output from the output end of the reducer 16. The output end of the reducer 16 is connected to the drive output shaft 17.

[0085] One end of the drive output shaft 17 is connected to the output end of the reducer 16, and the other end of the drive output shaft 17 is connected to the drive wheel of the robot body 1. The drive output shaft 17 is used to transmit the low-speed, high-torque rotary motion transmitted from the output end of the reducer 16 to the drive wheel, driving the robot body 1 to move along the inspection track.

[0086] Encoder 18 is fixed to the output end of reducer 16. When the output shaft of reducer 16 rotates, the rotating part of encoder 18 rotates synchronously with the output shaft, while the stationary part of encoder 18 remains stationary. Encoder 18 detects the rotation angle and number of rotations of the output shaft of reducer 16 and converts them into electrical pulse signals for output. Since the output end of reducer 16 is fixedly connected to drive output shaft 17, the rotation angle and number of rotations detected by encoder 18 correspond to the rotation angle and number of rotations of drive output shaft 17, which in turn correspond to the rotation angle and number of rotations of drive wheel.

[0087] The host computer control cabinet 8 receives the pulse signal output by the encoder 18, and calculates the distance the drive wheel rolls on the inspection track based on the pulse signal count and the relevant parameters of the drive wheel, which is the displacement information of the robot body 1.

[0088] This technical solution achieves precise walking control of the robot body through a three-stage transmission structure of "servo motor → reducer → drive output shaft". The servo motor provides power, the reducer reduces the speed and increases the torque to meet the low-speed, high-torque walking requirements of the inspection robot, and the drive output shaft transmits power to the drive wheels. The encoder is fixed to the output end of the reducer and directly measures the actual rotation of the drive wheels, eliminating measurement errors caused by gaps and deformations in the transmission chain from the servo motor to the reducer output end. Compared to solutions that mount the encoder on the motor shaft, the encoder in this technical solution measures the actual rotation of the final drive wheel, providing more accurate and reliable displacement information and ensuring hardware support for positioning accuracy.

[0089] In other embodiments of this application, the robotic arm module includes a lifting module, a robotic arm, and a precision scanning camera. The lifting module is vertically mounted on the robot body, the robotic arm is mounted on the movable end of the lifting module, and the precision scanning camera is mounted on the end effector of the robotic arm. The lifting module is used to drive the robotic arm to perform height interpolation in the vertical direction according to the trajectory of the wheel axle height change, so that the precision scanning camera maintains a constant working distance from the target wheel axle. The robotic arm module is used to perform precision scanning inspection in the opposite direction to the linear scanning inspection.

[0090] like Figure 1 As shown, this embodiment includes three robotic arm modules (first robotic arm module 9, second robotic arm module 10, and third robotic arm module 11), arranged along the length of the robot body 1. Each robotic arm module has the same structure, including a lifting module, a robotic arm, and a precision scanning camera.

[0091] The lifting module is mounted on the top of the robot body 1. The lifting module includes a fixed end and a movable end. The fixed end of the lifting module is fixedly connected to the top surface of the robot body 1, and its installation direction is vertical. The movable end of the lifting module can move vertically up and down relative to the fixed end. The lifting module is driven by a servo motor, which receives control commands from the host computer control cabinet 8 and controls the movable end to move to the target height position according to the commands.

[0092] The robotic arm is mounted on the movable end of the lifting module. The base end of the robotic arm is fixedly connected to the movable end of the lifting module. The robotic arm can adjust its position and attitude in the horizontal direction and / or multiple degrees of freedom in space. The end effector of the robotic arm is an end effector.

[0093] The precision scanning camera is mounted on the end effector of the robotic arm. The precision scanning camera includes a first line scan camera 12 and a second line scan camera 13, both used to acquire high-definition images of the bogie axles and blind-scan components under the train. The first line scan camera 12 and the second line scan camera 13 can be mounted on the end effectors of different robotic arm modules. For example, the first line scan camera 12 is mounted on the end effector of the first robotic arm module 9, and the second line scan camera 13 is mounted on the end effector of the second robotic arm module 10, thereby achieving multi-point parallel precision scanning operations. Both the first line scan camera 12 and the second line scan camera 13 are high-resolution line scan cameras. By moving along a preset path driven by the robotic arm, they perform line-by-line scanning imaging of key components under the train to obtain high-definition two-dimensional images. Figure 1 As shown in the example, the first line scan camera 12 is mounted on the end effector of the first robotic arm module 9, and the second line scan camera 13 is mounted on the end effector of the second robotic arm module 10.

[0094] During operation, the host computer control cabinet 8 sends height control commands to the lifting module based on the generated wheel axle height change trajectory. The wheel axle height change trajectory records the vertical height value of each target wheel axle. The host computer control cabinet 8 compares the height value of the target wheel axle with the preset standard working distance and calculates the height difference that the lifting module needs to compensate for. The servo motor of the lifting module drives the movable end to move vertically according to this difference, driving the robotic arm and the precision scanning camera (i.e., the first line scanning camera 12 or the second line scanning camera 13) to move up and down synchronously, so that the precision scanning camera maintains a constant working distance from the target wheel axle.

[0095] Meanwhile, the movement direction of the robotic arm module during fine scanning inspection is opposite to that during line scanning inspection. During line scanning inspection, the robot body 1 moves from the front to the rear of the vehicle. During fine scanning inspection, the robot body 1 moves from the rear to the front of the vehicle, and the robotic arm module performs fine scanning in the opposite direction. Each robotic arm module moves sequentially to the target wheel axle position according to the position point sequence in the wheel axle positioning trajectory, and the first line scanning camera 12 and the second line scanning camera 13 perform fine scanning to acquire images.

[0096] This application's technical solution constructs a precision scanning execution mechanism with height compensation function by sequentially installing a lifting module, a robotic arm, and a precision scanning camera. The lifting module actively interpolates in the vertical direction according to the trajectory of axle height changes, ensuring that the precision scanning camera maintains a constant working distance with axles at different heights. This guarantees the clarity and consistency of the precision scanned image throughout the inspection process and avoids image defocusing caused by axle height changes due to wheel rim wear. Simultaneously, the movement direction of the precision scanning inspection is opposite to that of the line scanning inspection, allowing the robotic arm to operate sequentially from the far end of the trajectory to the near end, avoiding interference from the already scanned area on the robotic arm's movement. The combination of these two aspects achieves safe, clear, and efficient fixed-point precision scanning.

[0097] In other embodiments of this application, the host computer control cabinet is used to establish a global coordinate system with the positioning starting point as the origin, generate an absolute position trajectory of the wheel axle positioning trajectory and the wheel axle height change trajectory with reference to the global coordinate system, and associate and store the absolute position trajectory with the train vehicle identification number.

[0098] When the host computer control cabinet 8 identifies the reference features at the bottom of the vehicle's front end using the point laser distance sensor, it records the current encoder 18 value as the positioning starting point. The host computer control cabinet 8 establishes a global coordinate system with this positioning starting point as the origin. Specifically, the host computer control cabinet 8 defines the horizontal position of this positioning starting point as the origin and the forward direction of the robot body 1 as the positive direction of the coordinate axes. During subsequent movement, the horizontal coordinate of each position is converted into the actual running distance relative to the origin using the displacement information from the encoder 18.

[0099] During the entire line scan motion, the host computer control cabinet 8, based on the fusion processing results of each wheel axle and the corresponding encoder displacement information, uses the global coordinate system as a reference to record the horizontal position of each wheel axle as the abscissa value and the vertical height of each wheel axle as the ordinate value, thereby generating wheel axle positioning trajectories (a sequence of horizontal positions of each wheel axle in the global coordinate system) and wheel axle height change trajectories (a sequence of vertical heights of each wheel axle in the global coordinate system). Both of these trajectories are absolute position trajectories and do not depend on the current position of the robot body 1.

[0100] After the trajectory is generated, the host computer control cabinet 8 obtains the train vehicle identification number (vehicle number) of the currently inspected train. The host computer control cabinet 8 associates the generated absolute position trajectory (including wheel axle positioning trajectory and wheel axle height change trajectory) with the train vehicle identification number and stores it in the storage medium of the host computer control cabinet 8 (such as hard disk, solid-state drive or cloud storage).

[0101] When the same train enters the inspection station again, the host computer control cabinet 8 retrieves the corresponding absolute position trajectory from the storage medium based on the train's vehicle identification number. The robot body 1 directly controls the traveling drive mechanism 14 to move to each wheel axle position according to the stored wheel axle positioning trajectory, and controls the robotic arm module to perform height compensation according to the stored wheel axle height change trajectory, performing fine sweeping action without having to repeat the line sweeping and track establishment process.

[0102] This technical solution establishes a global coordinate system with the starting point of the train's front as the origin, generating absolute position trajectories by mapping the horizontal and vertical positions of each axle to the vehicle identification number. The global coordinate system provides a fixed reference frame for the trajectory, ensuring it is independent of the robot's current docking position. After binding the trajectory to the vehicle, when the same vehicle re-enters the depot, the host computer can directly retrieve the existing trajectory based on the vehicle identification number, eliminating the need to re-execute the line-scanning and track-building process, and directly driving the traveling mechanism and robotic arm to perform fine scanning. This achieves a highly efficient inspection mode of "one-time positioning, repeated maintenance," significantly reducing the time spent on repeated inspections of the same type of EMU and improving operational efficiency.

[0103] 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.

[0104] 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 high-speed train axle detection, inspection, and positioning system, characterized in that, include: The robot itself is movably mounted on the inspection track at the bottom of the train. The sensing module, installed on the robot body, includes: At least two 3D line laser sensors are configured to scan the outline data of the underside of the train vertically upwards during travel; At least two point laser distance sensors are configured to measure distance data to various feature points on the bottom of the train vertically upwards; A walking drive mechanism is configured to drive the robot body to walk along the inspection track. An encoder is installed at the output end of the walking drive mechanism and is configured to acquire the displacement information of the robot body in real time. The host computer control cabinet is communicatively connected to the 3D line laser sensor, the point laser distance sensor and the encoder respectively; At least one robotic arm module is installed on the robot body and connected to the host computer control cabinet for performing fine cleaning actions; The host computer control cabinet is configured to identify the reference features at the bottom of the vehicle front through the point laser distance sensor and record the current value of the encoder as the positioning starting point. During the movement, the contour data collected by the 3D line laser sensor and the distance data collected by the point laser distance sensor are received simultaneously, and the displacement information of the encoder is correlated in real time. The contour data and the distance data are fused together. Based on the fusion result and the displacement information, the horizontal position and vertical height of each bogie wheel axle are calculated to generate wheel axle positioning trajectory and wheel axle height change trajectory. The robot arm module is moved to the target wheel axle position by controlling the traveling drive mechanism according to the wheel axle positioning trajectory, and the robot arm module is height compensated according to the wheel axle height change trajectory to perform fine sweeping action.

2. The high-speed train axle detection, inspection, and positioning system according to claim 1, characterized in that, The step of identifying the reference features at the bottom of the vehicle front using the point laser distance sensor and recording the current encoder value as the positioning starting point includes: The height information of the front of the vehicle is obtained based on the point laser distance sensor; The height information is compared with a preset threshold. The vehicle's front position is identified based on the comparison between the height information and a preset threshold, and the current encoder value is recorded. The current encoder value is converted into the position of the robot body on the inspection track as the positioning starting point.

3. The high-speed train axle detection, inspection, and positioning system according to claim 1, characterized in that, Two point laser distance sensors are arranged side by side along a direction perpendicular to the forward movement of the robot body. The distance data includes characteristic data of the bogie wheel axle and interference characteristic data similar to the bogie wheel axle. One point laser distance sensor is configured to collect the characteristic data of the bogie wheel axle, and the other point laser distance sensor is configured to collect interference characteristic data similar to the bogie wheel axle.

4. The high-speed train axle detection, inspection, and positioning system according to claim 3, characterized in that, The process of fusing the contour data and the distance data, calculating the horizontal position and vertical height of each bogie axle based on the fusion result and the displacement information, and generating axle positioning trajectories and axle height change trajectories includes: The characteristic data of the bogie wheel axle are compared with the interference characteristic data; Based on the comparison results, the interference features corresponding to the interference feature data are eliminated, and the true feature data of the bogie wheel axle is obtained. The horizontal position and vertical height of each bogie axle are calculated based on the actual axle feature data and the displacement information.

5. The high-speed train axle detection, inspection, and positioning system according to claim 4, characterized in that, The step of comparing the feature data of the bogie wheel axle with the interference feature data, eliminating the interference features corresponding to the interference feature data based on the comparison result, and obtaining the true feature data of the bogie wheel axle includes: The host computer control cabinet marks the arc features in the contour data collected by the 3D line laser sensor as candidate wheel axle positions; The host computer control cabinet performs spatial matching between the distance data collected by the point laser distance sensor and the position of the candidate wheel axle; The host computer control cabinet eliminates the interference features corresponding to the interference feature data based on the spatial matching results, and obtains the true feature data of the bogie wheel axle.

6. The high-speed train axle detection, inspection, and positioning system according to claim 1, characterized in that, The process of fusing the contour data and the distance data, calculating the horizontal position and vertical height of each bogie axle based on the fusion result and the displacement information, and generating axle positioning trajectories and axle height change trajectories includes: The encoder values ​​are acquired in real time; The encoder value is converted into the actual running distance of the robot body on the inspection track; Based on the fusion processing results and the actual running distance, the horizontal position and vertical height of each bogie wheel axle are calculated, and wheel axle positioning trajectory and wheel axle height change trajectory are generated.

7. The high-speed train axle detection, inspection, and positioning system according to claim 1, characterized in that, The scanning direction of the 3D line laser sensor is perpendicular to the forward direction of the robot body, the measurement direction of the point laser distance sensor is parallel to the scanning direction of the 3D line laser sensor, and the measurement point of the point laser distance sensor is located on or near the scanning line of the 3D line laser sensor.

8. The high-speed train axle detection, inspection, and positioning system according to claim 1, characterized in that, The walking drive mechanism includes a servo motor, a reducer, and a drive output shaft. The output end of the servo motor is connected to the input end of the reducer, and the output end of the reducer is connected to the drive output shaft. The drive output shaft is connected to the robot body and is configured to drive the robot body to walk. The encoder is fixed to the output end of the reducer and is configured to detect the rotation angle and number of rotations of the drive output shaft to obtain the displacement information.

9. The high-speed train axle detection, inspection, and positioning system according to claim 1, characterized in that, The robotic arm module includes a lifting module, a robotic arm, and a precision scanning camera. The lifting module is vertically mounted on the robot body, the robotic arm is mounted on the movable end of the lifting module, and the precision scanning camera is mounted on the end effector of the robotic arm. The lifting module is configured to drive the robotic arm to perform height interpolation in the vertical direction according to the trajectory of the wheel axle height change, so that the precision scanning camera maintains a constant working distance from the target wheel axle. The robotic arm module is configured such that the direction of movement of the robot body during precision scanning inspection is opposite to the direction of movement of the robot body during line scanning inspection.

10. The high-speed train axle detection, inspection, and positioning system according to claim 1, characterized in that, The host computer control cabinet is configured to establish a global coordinate system with the positioning starting point as the origin, generate the absolute position trajectory of the wheel axle positioning trajectory and the wheel axle height change trajectory with reference to the global coordinate system, and associate and store the absolute position trajectory with the train vehicle identification number.

Images