Brake shoe imaging regulation method and system, and storage medium

Abstract

The application provides a brake shoe imaging regulation method and system and a storage medium. The method comprises: acquiring a real-time speed of a train; inputting the real-time speed into a preset bogie dynamics model to predict an attitude angle of a target bogie corresponding to a target brake shoe under the real-time speed through the bogie dynamics model, the bogie dynamics model being constructed based on secondary suspension system parameters of a corresponding train model, wheel-rail contact geometric parameters and train dynamics equations; generating a space pose adjustment instruction for an imaging holder based on the attitude angle, the space pose adjustment instruction being used to drive the imaging holder to adjust an optical axis direction to compensate for an imaging plane deflection of the target brake shoe caused by a bogie attitude change; controlling the imaging holder to move according to the space pose adjustment instruction, and controlling a high-speed imaging device on the imaging holder to image the target brake shoe in response to the imaging holder moving to a target pose. The application realizes stable and clear imaging of the brake shoe in a high dynamic scene.

Description

Technical Field

[0001] This application relates to the field of imaging control technology, and in particular to a brake shoe imaging control method, system and storage medium. Background Technology

[0002] During high-speed operation of railway freight cars, brake shoes, as a crucial component of the braking system, directly affect the braking performance and operational safety of the train due to their wear condition. In railway freight car maintenance and safety inspection scenarios, traditional manual inspection methods for brake shoe condition suffer from low efficiency, high cost, and numerous safety hazards. In recent years, with the development of trackside visual inspection technology, more and more systems are attempting to use high-speed imaging equipment for non-contact image acquisition and intelligent recognition of brake shoes during train operation, enabling real-time inspection even when the train is not stopped.

[0003] In related technologies, visual inspection of brake shoes mainly employs a fixed-angle gimbal control scheme. This involves pre-setting the pitch and yaw angles of the imaging gimbal to maintain a fixed imaging optical axis, and then acquiring images of the brake shoes using a high-speed imaging device mounted on the gimbal. However, in practical applications, due to significant variations in train speed, and the relatively low position of the brake shoes within the train body and their mounting on the bogie, they are susceptible to dynamic attitude changes such as vehicle pitch and roll. This makes it difficult to accurately align the imaging optical axis with the brake shoe imaging plane, resulting in image skew, blurring, or target loss. These issues severely restrict image quality and subsequent recognition accuracy, ultimately affecting the brake shoe status detection results.

[0004] Therefore, there is an urgent need for an effective brake shoe imaging control scheme to achieve stable and clear imaging of the brake shoe in high dynamic scenarios. Summary of the Invention

[0005] This application provides a brake shoe imaging control method, system, and storage medium to solve the problem in related technologies where the imaging optical axis is difficult to accurately align with the brake shoe imaging plane, resulting in problems such as image skew, blurring, or target loss, which seriously restrict image quality and subsequent recognition accuracy, and affect the brake shoe status detection results.

[0006] In a first aspect, this application provides a brake shoe imaging control method, comprising:

[0007] Obtain the train's real-time speed;

[0008] The real-time speed is input into the preset bogie dynamics model, and the attitude angle of the bogie corresponding to the target brake shoe at the real-time speed is predicted by the bogie dynamics model. The bogie dynamics model is constructed based on the secondary suspension system parameters, wheel-rail contact geometry parameters and train dynamics equations of the corresponding train model.

[0009] Based on the attitude angle, a spatial pose adjustment command for the imaging gimbal is generated. The spatial pose adjustment command is used to drive the imaging gimbal to adjust the optical axis direction in order to compensate for the deflection of the imaging surface caused by the bogie attitude change of the target brake shoe.

[0010] According to the spatial pose adjustment command, control the movement of the imaging gimbal, and in response to the movement of the imaging gimbal to the target pose, control the high-speed imaging device on the imaging gimbal to image the target brake shoe.

[0011] In one possible implementation, based on the attitude angle, spatial pose adjustment commands for the imaging gimbal are generated, including:

[0012] Based on the geometric parameters of the target brake shoe's installation position on the corresponding bogie, the spatial deflection direction of the normal vector corresponding to the target brake shoe's imaging surface is determined according to the attitude angle.

[0013] By applying the spatial vector inverse solution algorithm, the spatial coordinate inverse solution of the spatial deflection direction is obtained to obtain the target pitch angle and target yaw angle of the imaging gimbal.

[0014] Based on the target pitch angle and target yaw angle, spatial pose adjustment commands for the imaging gimbal are generated.

[0015] In one possible implementation, based on the target pitch angle and the target yaw angle, a spatial pose adjustment command for the imaging gimbal is generated, including:

[0016] Obtain the initial pitch and initial yaw angles of the imaging gimbal;

[0017] Determine the first pitch angle deviation of the target pitch angle relative to the initial pitch angle, and determine the first yaw angle deviation of the target yaw angle relative to the initial yaw angle;

[0018] Based on the built-in PID controller, a spatial pose adjustment command is generated to adjust the spatial pose of the imaging gimbal according to the first pitch angle deviation and the first yaw angle deviation.

[0019] In one possible implementation, the spatial pose adjustment command carries the target pitch angle and the target yaw angle, and the imaging gimbal is moved to the target pose in the following way:

[0020] Obtain the actual pitch angle and actual yaw angle of the imaging gimbal during its movement;

[0021] Determine the second pitch angle deviation between the actual pitch angle and the target pitch angle;

[0022] Determine the second yaw angle deviation between the actual yaw angle and the target yaw angle;

[0023] If the second pitch angle deviation and the second yaw angle deviation meet the pose shift conditions, then the imaging gimbal is determined to have moved to the target pose. The pose shift conditions include that the second pitch angle deviation is less than the pitch angle deviation threshold and the second yaw angle deviation is less than the yaw angle deviation threshold.

[0024] In one possible implementation, determining the motion of the imaging gimbal to the target pose includes:

[0025] If, within the set duration, both the second pitch angle deviation and the second yaw angle deviation meet the pose positioning conditions, then the imaging gimbal is determined to have moved to the target pose.

[0026] In one possible implementation, controlling a high-speed imaging device on an imaging pan-tilt unit to image the target brake shoe includes:

[0027] Based on the mapping relationship between train speed and exposure time, the target exposure time is determined according to the real-time speed;

[0028] Based on the target exposure time, generate imaging acquisition instructions;

[0029] Send an imaging acquisition command to the high-speed imaging device to trigger the high-speed imaging device to image the target brake shoe.

[0030] In one possible implementation, the brake shoe imaging control method further includes:

[0031] Acquire brake shoe images of the target brake shoe output by a high-speed imaging device;

[0032] The brake shoe image is associated with real-time speed, target pitch angle, target yaw angle, and imaging timestamp, and the associated and labeled brake shoe image is output.

[0033] In one possible implementation, controlling the movement of the imaging gimbal according to a spatial pose adjustment command includes:

[0034] Send spatial pose adjustment commands to the servo driver corresponding to the imaging gimbal to drive the servo motor of the imaging gimbal to adjust the pose of the imaging gimbal along the pitch and yaw axes.

[0035] Secondly, this application provides a brake shoe imaging control system, comprising:

[0036] An imaging control unit is used to execute the brake shoe imaging control method as described in the first aspect above;

[0037] The imaging gimbal is connected in communication with the imaging control unit. Under the control of the imaging control unit, it is used to adjust the spatial pose to adjust the direction of the optical axis in order to compensate for the deflection of the imaging surface caused by the bogie attitude change of the target brake shoe.

[0038] The high-speed imaging device is mounted on the imaging pan-tilt unit and communicates with the imaging control unit. The high-speed imaging device is used to image the target brake shoe under the control of the imaging control unit.

[0039] Thirdly, this application provides a brake shoe imaging control device, applied to an imaging control unit, the brake shoe imaging control device comprising:

[0040] The acquisition module is used to acquire the real-time speed of the train;

[0041] The input module is used to input the real-time speed into the preset bogie dynamics model. The bogie dynamics model predicts the attitude angle of the bogie corresponding to the target brake shoe at the real-time speed. The bogie dynamics model is constructed based on the secondary suspension system parameters, wheel-rail contact geometry parameters and train dynamics equations of the corresponding train model.

[0042] The generation module is used to generate spatial pose adjustment commands for the imaging gimbal based on the attitude angle. The spatial pose adjustment commands are used to drive the imaging gimbal to adjust the optical axis direction to compensate for the deflection of the imaging surface caused by the bogie attitude change of the target brake shoe.

[0043] The control module is used to control the movement of the imaging gimbal according to the spatial pose adjustment command, and in response to the movement of the imaging gimbal to the target pose, control the high-speed imaging device on the imaging gimbal to image the target brake shoe.

[0044] In one possible implementation, the generation module is specifically used to: determine the spatial deflection direction of the normal vector corresponding to the imaging surface of the target brake shoe based on the geometric parameters of the installation position of the target brake shoe on the corresponding bogie, according to the attitude angle; apply the spatial vector inverse solution algorithm to perform spatial coordinate inverse solution on the spatial deflection direction to obtain the target pitch angle and target yaw angle of the imaging gimbal; and generate spatial pose adjustment commands for the imaging gimbal based on the target pitch angle and target yaw angle.

[0045] In one possible implementation, the generation module is further configured to: acquire the initial pitch angle and initial yaw angle of the imaging gimbal; determine the first pitch angle deviation of the target pitch angle relative to the initial pitch angle, and determine the first yaw angle deviation of the target yaw angle relative to the initial yaw angle; and, based on the built-in PID controller, generate a spatial pose adjustment command for adjusting the spatial pose of the imaging gimbal according to the first pitch angle deviation and the first yaw angle deviation.

[0046] In one possible implementation, the spatial pose adjustment command carries the target pitch angle and the target yaw angle. The imaging gimbal is moved to the target pose by: acquiring the actual pitch angle and the actual yaw angle of the imaging gimbal during the movement; determining the second pitch angle deviation between the actual pitch angle and the target pitch angle; determining the second yaw angle deviation between the actual yaw angle and the target yaw angle; if the second pitch angle deviation and the second yaw angle deviation meet the pose move-to-position conditions, then the imaging gimbal is determined to have moved to the target pose. The pose move-to-position conditions include that the second pitch angle deviation is less than the pitch angle deviation threshold and the second yaw angle deviation is less than the yaw angle deviation threshold.

[0047] In one possible implementation, determining that the imaging gimbal has moved to the target pose includes: if, within a set duration, both the second pitch angle deviation and the second yaw angle deviation satisfy the pose movement into position condition, then the imaging gimbal has moved to the target pose.

[0048] In one possible implementation, the control module is specifically used to: determine the target exposure time based on the real-time speed according to the mapping relationship between train speed and exposure time; generate an imaging acquisition command based on the target exposure time; and send the imaging acquisition command to the high-speed imaging device to trigger the high-speed imaging device to image the target brake shoe.

[0049] In one possible implementation, the brake shoe imaging control device further includes an output module, which is used to acquire the brake shoe image of the target brake shoe output by the high-speed imaging device; associate the brake shoe image with the real-time speed, target pitch angle, target yaw angle, and imaging timestamp, and output the associated and labeled brake shoe image.

[0050] In one possible implementation, the control module is further configured to: send a spatial pose adjustment command to the servo driver corresponding to the imaging gimbal, so as to drive the servo motor of the imaging gimbal to adjust the pose of the imaging gimbal along the pitch axis and yaw axis.

[0051] Fourthly, this application provides an imaging control unit, including: a memory and a processor;

[0052] The memory stores the instructions that the computer executes;

[0053] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0054] Fifthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible embodiments of the first aspect.

[0055] In a sixth aspect, this application provides a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0056] The brake shoe imaging control method, system, and storage medium provided in this application acquire the real-time speed of the train and dynamically predict the attitude angle of the bogie corresponding to the target brake shoe at the real-time speed based on the bogie dynamics model. The bogie dynamics model is constructed based on the secondary suspension system parameters, wheel-rail contact geometry parameters, and train dynamics equations of the corresponding train model, thereby achieving accurate and pre-quantification of bogie attitude changes. Furthermore, it proactively performs feedforward correction of the imaging optical axis direction before imaging the target brake shoe, effectively avoiding the perspective shift problem caused by changes in train speed or track inequality, improving the spatial alignment accuracy of the imaging perspective, and ensuring imaging quality. Moreover, when the imaging gimbal is detected to have moved to the target pose, the high-speed imaging device on the imaging gimbal is triggered to image the target brake shoe. This timing of the imaging operation effectively avoids the problem of starting exposure while the imaging gimbal is still in the process of dynamic adjustment, ensuring the clarity and stability of each frame of brake shoe image, which is particularly suitable for high-speed railway freight car scenarios. Ultimately, this application achieves stable and clear imaging of brake shoes in high-dynamic scenarios, improving the accuracy of subsequent brake shoe state detection results. Attached Figure Description

[0057] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0058] Figure 1 A schematic diagram of a scenario for the brake shoe imaging control method provided in an embodiment of this application;

[0059] Figure 2 A schematic flowchart of a brake shoe imaging control method provided in an embodiment of this application;

[0060] Figure 3 A schematic flowchart of a brake shoe imaging control method provided in another embodiment of this application;

[0061] Figure 4 This is a schematic diagram of the structure of a brake shoe imaging control device provided in an embodiment of this application;

[0062] Figure 5 This is a schematic diagram of the structure of a brake shoe imaging control device provided in another embodiment of this application;

[0063] Figure 6 This is a schematic diagram of the imaging control unit provided in an embodiment of this application.

[0064] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0065] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0066] The terms “first,” “second,” etc., used in the specification and claims of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, products, or apparatus.

[0067] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0068] In related technologies, the fixed-angle gimbal control scheme is simple and easy to implement, but it cannot adapt to the dynamic attitude changes of the bogie caused by the change of train running speed. This results in an angle between the imaging optical axis and the normal vector of the brake shoe imaging surface, making it difficult to guarantee image quality. Problems such as image skew, blurring, or target loss may occur, which in turn affects the brake shoe status detection results.

[0069] To improve image quality, some related technologies employ post-processing correction schemes, which involve using software algorithms to correct distortion or compensate for blur after image acquisition. These methods rely on the robustness of the image processing algorithms, but in high-speed scenarios where image blur is severe, post-processing struggles to recover clear details and cannot address the fundamental problem of imaging viewpoint shift. Furthermore, they suffer from response lag.

[0070] In addition, there is a common problem in the timing control of imaging gimbal adjustment and imaging exposure triggering in related technologies. That is, the exposure is started before the imaging gimbal has stabilized in place, which further leads to a decrease in image quality.

[0071] To address the aforementioned technical problems, the inventors, starting from practical testing needs, first analyzed the root cause of brake shoe imaging blurring during high-speed operation: the dynamic attitude changes of the bogie leading to the offset of the imaging optical axis. To solve this problem, they proposed introducing a bogie dynamics model, combining train speed with secondary suspension system parameters and wheel-rail contact geometry parameters to dynamically predict the attitude angle. Subsequently, based on the attitude angle, they calculated the required adjustment angle of the imaging gimbal, and only triggered imaging after detecting the gimbal's positioning, avoiding false triggering during dynamic adjustments. Finally, through a collaborative design integrating speed perception, attitude prediction, imaging gimbal control, and imaging triggering, stable and clear imaging of the brake shoes in highly dynamic scenarios was achieved.

[0072] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0073] First, the application scenarios of this application will be introduced.

[0074] This application is applicable to the visual inspection scenario of railway freight cars along the track, such as Figure 1This is a schematic diagram of a scenario illustrating the brake shoe imaging control method provided in this application embodiment. The scenario includes an imaging control unit 11, an imaging pan-tilt unit 12, and a high-speed imaging device 13. The brake shoe imaging control system, comprised of the imaging control unit 11, imaging pan-tilt unit 12, and high-speed imaging device 13, is deployed at detection stations along the track. The brake shoe imaging control method provided in this application embodiment is run in the imaging control unit 11. For example, this scenario also includes multiple speed sensors deployed beside the track. When the train is running, the speed sensors collect wheel-passing pulse signals in real time. The imaging control unit 11 acquires these pulse signals and executes the brake shoe imaging control method, driving the imaging pan-tilt unit 12 to adjust the imaging optical axis direction. After the imaging pan-tilt unit is stably positioned, the imaging control unit 11 triggers the high-speed imaging device 13 to acquire an image of the target brake shoe. This application is suitable for scenarios with a wide train speed range (low to high speed) and complex operating environments (uneven tracks, curved sections), ensuring that brake shoe images can still be clearly acquired under dynamic attitude changes.

[0075] It should be noted that, Figure 1 The application scenarios shown are for illustrative purposes only. This application does not limit the specific form, implementation method, deployment architecture, or quantity of the imaging control unit 11, imaging gimbal 12, and high-speed imaging device 13. The imaging control unit 11 can be any electronic device or combination of devices with computing and control capabilities, including but not limited to: industrial control computers, embedded computers, servers, desktop computers, laptops, tablets, or server clusters and distributed computing systems composed of multiple servers. The high-speed imaging device 13 can be any device capable of responding to external trigger signals and completing image acquisition within a controllable exposure time. For example, it can be an industrial camera, scientific camera, high-speed camera using a global shutter or rolling shutter, or a smart camera integrating an image sensor and processor. The imaging gimbal 12 includes a precision servo mechanism with at least two rotational degrees of freedom (yaw and pitch) for supporting and precisely driving the high-speed imaging device 13. Its specific form can be a two-axis servo gimbal, a three-axis stabilized gimbal, or integrated into a robot end effector.

[0076] In terms of quantity and deployment, the brake shoe imaging control system can include one or more imaging control units 11, which work together in a centralized or distributed manner. Similarly, one or more imaging subsystems consisting of imaging pan-tilt units 12 and high-speed imaging devices 13 can be deployed to simultaneously acquire images of one or more target points (such as the left and right brake shoes of different bogies) of the same train in a synchronous or asynchronous manner. Multiple subsystems can be coordinated by the same control unit or different control units.

[0077] Figure 2 This is a schematic flowchart of a brake shoe imaging control method provided in an embodiment of this application, as shown below. Figure 2As shown, the brake shoe imaging control method includes:

[0078] S201. Obtain the real-time speed of the train.

[0079] For example, in one implementation, multiple speed sensors deployed along the track collect real-time pulse signals from passing train wheels. The real-time train speed is then calculated based on a fixed spacing and time difference. Specifically, at least two speed sensors are fixedly deployed on each side of the track. When a train wheel passes a speed sensor, a pulse signal is triggered, and the real-time train speed is calculated using the known fixed spacing and the measured time difference. , represented as:

[0080]

[0081] In the formula, For the train at the time The real-time speed of the track crossing, in units of ; The fixed spacing between adjacent speed sensors on one side of the rail, in units of . The value range is 2. -10 This range of values ​​ensures that a significant time difference is formed during the passage of the wheel, but it should not be too long to avoid amplifying the error caused by speed changes. This represents the time difference between the passing of two adjacent speed sensors, expressed in units of [unit missing]. The value ranges from 0.08 to 1.5. It can be acquired with microsecond-level time accuracy by high-speed acquisition equipment.

[0082] By forming multiple speed measurement sections with each set of speed measurement pairs, dynamic tracking of train speed at different times or locations can be achieved.

[0083] For example, in another implementation, the imaging control unit 11 receives data packets sent by the train's onboard system via wireless communication. The data packets carry the real-time speed, and the real-time speed is obtained by parsing the data packets. The train's onboard system is, for example, a train control and management system or a train automatic protection system.

[0084] S202. Input the real-time speed into the preset bogie dynamics model, and predict the attitude angle of the bogie corresponding to the target brake shoe at the real-time speed through the bogie dynamics model. The bogie dynamics model is constructed based on the secondary suspension system parameters, wheel-rail contact geometry parameters and train dynamics equations of the corresponding train model.

[0085] The bogie dynamics model is essentially a mathematical model that establishes a functional mapping between real-time speed and attitude angle.

[0086] For example, the result obtained in step S201 As the primary control variable, it is input into the bogie dynamics model. If the bogie attitude angle includes pitch angle... Roll angle The mapping relationship between real-time velocity and attitude angle is expressed as:

[0087]

[0088] In the formula, The total mass of the bogie is expressed in units of... The value range is 5000. -8000 ; The stiffness of the secondary suspension system is expressed in units of 1. The range of values ​​is It characterizes the stiffness of air springs or steel springs, and affects the vehicle body's response characteristics to disturbances; This is the damping coefficient of the secondary suspension system, in units of... The range of values ​​is Control the oscillation amplitude of the system to prevent drastic changes in attitude angle; These are the wheel-rail contact geometry parameters. This refers to the rim taper, with a value ranging from 0.03 to 0.07. It is a correction factor for track gauge difference or wheel-rail inclination angle, with a value range of 0.001-0.02, reflecting the correction effect of factors such as track gauge inconsistency and rail surface wear on the roll angle. This represents the response function established based on vehicle dynamics, specifically including:

[0089] (1) Pitch angle function Speed ​​of use Using as the independent variable, a polynomial relationship describing the bogie's pitch motion is fitted, expressed as the pitch angle function:

[0090] ;

[0091] in, The initial offset value of the bogie's inherent pitch angle at low speed is -1.0° to 1.0°, reflecting the slight pitch angle caused by the vehicle's center of gravity or installation error when the vehicle is static or running at low speed. The coefficient for the linear effect of speed ranges from -0.2 to 0.3, representing the linear trend of the effect of increased speed on the pitch angle. Positive values ​​indicate a pitching trend, while negative values ​​indicate a pitching trend. The coefficient of the quadratic term ranges from -0.01 to 0.02, representing the beginning of the nonlinear effect of speed and reflecting the combined effect of vehicle inertia and damping. The coefficient is a cubic term, ranging from -0.001 to 0.005. It is used to adjust the curvature of the high-speed segment, control the rise or fall of the fitting end, and prevent non-physical angle changes.

[0092] (2) Roll angle function Similarly, based on train speed As input, fit the corresponding lateral roll response, where the roll angle is expressed as:

[0093] ;

[0094] in, The initial roll offset angle of the bogie at low speed ranges from -0.5° to 0.5° and is caused by factors such as uneven rail surface and inconsistent initial attitude. This is the linear response coefficient of the roll angle to speed, with a value ranging from -0.05 to 0.1. It is generally smaller than the linear term of the pitch angle because roll is mainly driven by lateral disturbances. This is the coefficient of the quadratic term, with a value range of -0.002 to 0.008, reflecting the influence of lateral acceleration, which comes from the lateral slope of the track or unevenness of the track bed; The coefficient is a cubic term, ranging from -0.0005 to 0.003. It adjusts the minimum angle increase during high-speed operation, enhances fitting flexibility, and prevents overfitting.

[0095] Based on the relative position of the imaging point (which should be understood as a pre-set fixed spatial reference position, representing the point where the imaging optical axis is exactly aligned with the brake shoe imaging plane) and the bogie containing the target brake shoe on the train, i.e., the relative distance between the center position of the bogie and the imaging point. Calculate the correlation time between the imaging point and the bogie at the current moment. ,satisfy Predicting imaging time Substituting these values ​​into the bogie dynamics model, we can obtain the attitude angle at the corresponding moment, expressed as: .in, The relative distance between the imaging point and the center of the target bogie, in units of . The value range is -12. Up to 12 The truck's body length is approximately 13. -14 The brake shoes and bogies have a fixed relative positional difference, which can be represented by positive and negative signs to indicate their front and rear positional differences. The predicted imaging time associated with the imaging point and the bogie, in units of ; This indicates the pitch angle of the bogie at the moment of predicted imaging. This represents the roll angle of the bogie at the moment of predicted imaging. The aforementioned attitude angle constitutes the dynamic attitude angle input for subsequent imaging compensation control.

[0096] S203. Based on the attitude angle, generate a spatial pose adjustment command for the imaging gimbal. The spatial pose adjustment command is used to drive the imaging gimbal to adjust the optical axis direction to compensate for the deflection of the imaging surface caused by the bogie attitude change of the target brake shoe.

[0097] It should be understood that this attitude angle is the bogie's attitude angle at the moment of predicted imaging, and also reflects the actual attitude angle of the bogie at the real-time speed during train operation.

[0098] Brake shoes are mounted on bogies. As the train speed changes, the bogies will experience pitch and roll changes, which in turn causes the imaging surface of the brake shoes fixed on them to deflect in space (i.e., deviate from the ideal imaging surface of the brake shoes when the train is stationary). Without compensation, an angle will appear between the imaging optical axis and the normal vector of the brake shoe surface, resulting in image blurring, distortion, or viewing angle shift.

[0099] For example, this step uses the bogie's attitude angle to calculate the attitude angle to be adjusted of the imaging gimbal, and then generates a spatial pose adjustment command to control the imaging gimbal to change its pose, thereby adjusting the direction of the optical axis so that the imaging optical axis is aligned with the brake shoe imaging plane, that is, the angle between the imaging optical axis and the brake shoe surface normal vector is 0.

[0100] S204. According to the spatial pose adjustment command, control the movement of the imaging gimbal, and in response to the movement of the imaging gimbal to the target pose, control the high-speed imaging device on the imaging gimbal to image the target brake shoe.

[0101] For example, the imaging control unit 11 sends corresponding control signals (i.e., spatial pose adjustment commands) to the servo driver, stepper driver, direct drive motor controller or piezoelectric driver of the imaging gimbal, or sends spatial pose adjustment commands to the local intelligent controller built into the imaging gimbal, driving the imaging gimbal to move to the target pose.

[0102] Specifically, in one implementation, controlling the movement of the imaging gimbal according to the spatial pose adjustment command includes: sending the spatial pose adjustment command to the servo driver corresponding to the imaging gimbal to drive the servo motor of the imaging gimbal to adjust the pose of the imaging gimbal along the pitch axis and yaw axis.

[0103] For example, in one implementation, when controlling the movement of the imaging gimbal, the imaging control unit 11 monitors in real time the position lock signal sent by the imaging gimbal side. This position lock signal indicates that the imaging gimbal has reached the target pose. Therefore, when the imaging control unit 11 detects this position lock signal, it determines that the imaging gimbal has moved to the target pose, and at this time, it controls the high-speed imaging device to acquire images of the target brake shoe.

[0104] In another implementation, the imaging control unit 11 acquires the actual attitude angle of the imaging gimbal in real time, and then uses a preset algorithm to determine whether the imaging gimbal has moved to the target pose based on the actual attitude angle.

[0105] In this embodiment, by acquiring the real-time speed of the train, the attitude angle of the bogie corresponding to the target brake shoe at the real-time speed is dynamically predicted based on the bogie dynamics model. The bogie dynamics model is constructed based on the secondary suspension system parameters, wheel-rail contact geometry parameters, and train dynamics equations of the corresponding train model. This enables accurate and advance quantification of bogie attitude changes, and thus proactively performs feedforward correction of the imaging optical axis direction before imaging the target brake shoe. This effectively avoids the perspective shift problem caused by changes in train speed or track inequality, improves the spatial alignment accuracy of the imaging perspective, and ensures imaging quality. Furthermore, when the imaging gimbal is detected to have moved to the target pose, the high-speed imaging device on the imaging gimbal is triggered to image the target brake shoe. This imaging timing effectively avoids the problem of starting exposure while the imaging gimbal is still in the process of dynamic adjustment, ensuring the clarity and stability of each frame of brake shoe image, which is particularly suitable for high-speed railway freight car scenarios. Ultimately, this application achieves stable and clear imaging of brake shoes in high-dynamic scenarios, improving the accuracy of subsequent brake shoe state detection results.

[0106] In some embodiments, spatial pose adjustment commands for the imaging gimbal are generated based on the attitude angle, specifically including:

[0107] S2031. Based on the geometric parameters of the target brake shoe's installation position on the corresponding bogie, determine the spatial deflection direction of the normal vector corresponding to the target brake shoe's imaging surface according to the attitude angle.

[0108] In this embodiment, the installation position geometric parameters represent the ideal situation (i.e., the train is stationary and the bogie has not deflected), when the target brake shoe is installed on the bogie, and the reference normal vector of the target brake shoe imaging surface is along the direction vertically upward along the car body.

[0109] For example, the reference normal vector of the target brake shoe imaging surface. Considering that the bogie generates a pitch angle during train operation With roll angle Then, the normal vector of the brake shoe imaging surface undergoes a composite rotation relative to the static coordinate system. Using the Euler angle combination rotation matrix, the normal vector under dynamic attitude can be calculated. The normal vector of the brake shoe imaging surface under dynamic conditions is expressed as: .

[0110] In the formula, The pitch angle of the bogie is typically within the range of ±0.01 to ±0.1 rad. During high-speed operation, the longitudinal attitude deflection caused by road disturbance or load inertia can be covered by this range, which can cover medium and high-speed freight conditions. The bogie roll angle typically ranges from ±0.005 to ±0.05 rad. It is the lateral roll of the train caused by lateral disturbances or eccentric loading, and is usually less than the pitch angle. , represents the rotation matrix about the X-axis. , which represents the rotation matrix about the Y-axis.

[0111] In other words, this step substitutes the bogie's pitch and roll angles into the aforementioned spatial rotation matrix, and then applies them to the target brake shoe's static normal vector. This allows for the calculation of the actual spatial orientation of the normal vector of the target brake shoe imaging surface under the current attitude, i.e., the spatial deflection direction. The spatial deflection result of the normal vector reflects the influence of bogie attitude changes on the imaging angle and provides a clear target direction for subsequent imaging optical axis adjustment. The overall design meets the technical requirements for dynamic visual compensation during train operation, ensuring that the imaging optical axis can be accurately aligned with the target brake shoe in real time, achieving clear imaging even at high speeds.

[0112] S2032. Apply the spatial vector inverse solution algorithm to perform spatial coordinate inverse solution on the spatial deflection direction to obtain the target pitch angle and target yaw angle of the imaging gimbal.

[0113] It should be understood that in order for the imaging optical axis of the imaging gimbal to be collinear with the normal vector of the target gate tile's imaging surface (i.e., the imaging optical axis to be aligned with the target gate tile), the orientation of the imaging optical axis of the imaging gimbal must be consistent with the direction of this normal vector. This can be achieved through inverse analysis of the spatial coordinate system to complete the attitude calculation.

[0114] For example, based on the theory that the optical axis of the imaging gimbal should be collinear with the normal vector of the target tile's imaging surface, the target attitude angle required by the imaging gimbal can be calculated through inverse spatial vector calculation. The target attitude angle includes the target pitch angle. and target yaw angle , represented as:

[0115]

[0116] in, Normal vector The three components, This is the pitch angle that the imaging gimbal should be adjusted to, in rad. The yaw angle that the imaging gimbal should be adjusted to, in rad; using Guarantee the uniqueness and continuity of attitude solutions in the four quadrants.

[0117] This step, after determining the actual spatial normal vector (the deflected normal vector) of the target brake tile's imaging surface, treats the deflected normal vector as the target direction vector. Combining this with the default initial optical axis direction, it uses inverse trigonometric functions to deduce the two angle values ​​that need adjustment: first, the pitch angle, i.e., how much the imaging gimbal needs to tilt upwards or downwards around the X-axis; and second, the yaw angle, i.e., how much the imaging gimbal needs to turn left or right around the vertical axis. The yaw angle is calculated using the four-quadrant arctan² function. This can effectively avoid jumps or non-uniqueness of solutions when the angle is close to ±180°.

[0118] This step calculates the specific spatial attitude angle that the imaging gimbal should rotate to, ensuring that the imaging optical axis can be accurately aligned with the current facing direction of the target brake shoe.

[0119] S2033. Based on the target pitch angle and target yaw angle, generate spatial pose adjustment commands for the imaging gimbal.

[0120] For example, the target pitch and yaw angles are first validated numerically to check for null values ​​or non-numerical types. Then, a physical range check is performed. For instance, the target yaw angle must be within a mechanical limit range of -180 degrees to +180 degrees, and the target pitch angle must be within a mechanical limit range of -90 degrees to +30 degrees. If these angles exceed the range, they are clamped to the allowable limits, and a warning log is generated. Further low-pass filtering can be performed. Finally, the filtered target pitch and yaw angles are encapsulated into a spatial pose adjustment command to generate a spatial pose adjustment command for the imaging gimbal. For example, if the imaging gimbal's servo driver receives a target pitch angle of 10° and a target yaw angle of 5°, it controls the servo motor to rotate, adjusting the imaging gimbal's pose to a pitch angle of 10° and a yaw angle of 5°. Alternatively, calculate the pitch angle difference between the filtered target pitch angle and the initial pitch angle before the imaging gimbal pose adjustment, and the yaw angle difference between the filtered target yaw angle and the initial yaw angle before the imaging gimbal pose adjustment. Then, encapsulate the pitch angle difference and yaw angle difference into a spatial pose adjustment command, thereby instructing the servo driver to control the servo motor to rotate the yaw angle difference and pitch angle difference. That is, the imaging gimbal is controlled to move to the target pitch angle and target yaw angle with only one spatial pose adjustment command.

[0121] In this embodiment, by introducing geometric parameters of the brake shoe installation position and inverse spatial coordinate solving technology, high-precision calculation of the target attitude angle of the imaging gimbal is achieved. Based on this high-precision target attitude angle, a spatial pose adjustment command is generated to ensure that the imaging optical axis after pose adjustment can be accurately aligned with the current facing direction of the target brake shoe.

[0122] Furthermore, in one possible implementation, generating spatial pose adjustment commands for the imaging gimbal based on the target pitch angle and target yaw angle may include:

[0123] S2033-1. Obtain the initial pitch angle and initial yaw angle of the imaging gimbal.

[0124] For example, the initial pitch angle and initial yaw angle of the imaging gimbal are acquired in real time by a high-precision encoder or an inertial measurement unit, and the imaging control unit 11 obtains the initial pitch angle and initial yaw angle of the imaging gimbal from the high-precision encoder or the inertial measurement unit.

[0125] In this step, the initial pitch angle should be understood as the pitch angle before each attitude adjustment. Therefore, the initial pitch angle and the initial yaw angle are variables.

[0126] S2033-2. Determine the first pitch angle deviation of the target pitch angle relative to the initial pitch angle, and determine the first yaw angle deviation of the target yaw angle relative to the initial yaw angle.

[0127] For example, the initial pitch angle The difference between the pitch angle and the target angle is calculated and expressed as: Initial yaw angle The difference between the yaw angle and the target yaw angle is calculated and expressed as: .in, The difference in the elevation angle adjustment of the imaging gimbal (first elevation angle deviation), in rad; The difference in yaw angle adjustment for the imaging gimbal (first yaw angle deviation), in rad; The target pitch angle of the imaging gimbal typically ranges from ±0.05 to ±0.25 rad, covering the optical axis alignment angle caused by the deflection of the brake shoes during high-speed operation; The target yaw angle of the imaging gimbal is typically within the range of ±0.05 to ±0.25 rad. It has a relatively small impact on lateral attitude changes, but larger angle compensation needs to be considered in curved sections or when the target is off track.

[0128] S2033-3: Based on the built-in PID controller, a spatial pose adjustment command is generated to adjust the spatial pose of the imaging gimbal according to the first pitch angle deviation and the first yaw angle deviation.

[0129] This step will , The angular displacement command is converted into the corresponding servo motor, and a low-latency drive signal is generated through the built-in PID controller to perform the rotation adjustment of the imaging gimbal, ensuring that the imaging optical axis is accurately aligned with the normal direction of the target brake shoe imaging surface.

[0130] It should be understood that, based on the first pitch angle deviation and the first yaw angle deviation, a sequence of spatial pose adjustment commands is generated to drive the motion of the imaging gimbal servo motors. .

[0131] For example, the spatial pose adjustment command can be represented as:

[0132] ;

[0133] ;

[0134] in, This indicates the pitch angle adjustment control signal. This indicates the yaw angle adjustment control signal, used to drive the servo motor; This is the proportional gain coefficient, with a value ranging from 5 to 20, used to increase the system response speed; This is the integral gain coefficient, with a value ranging from 0.1 to 1.0, used to eliminate steady-state error; This is the differential gain coefficient, with a value ranging from 0.01 to 0.5, which improves the system's ability to resist disturbances and suppresses oscillations. Indicates pitch angle error in time The integral term within the range represents the cumulative error, with the unit being rad·s and a value range of 0-5 rad·s. It is related to the duration and magnitude of the error and affects the steady-state response of the controller. This is the integral term for yaw angle error, with a value range of 0-5 rad·s and a limited integration amplitude.

[0135] All of these are PID coefficients, which are obtained using an offline tuning method. For example, by applying a step angle command to the imaging gimbal system on an experimental platform, recording its response curve, and extracting dynamic characteristic indicators such as overshoot, rise time, and oscillation period, the Ziegler-Nichols empirical tuning method is then used. By adjusting the proportional gain, the system reaches the critical oscillation state, the critical gain and oscillation period are determined, and appropriate proportional, integral, and derivative coefficients are calculated accordingly. Finally, the tuning results are imported into the PID controller for fine-tuning and optimization in actual measurements to ensure that the system has a fast response speed, small overshoot, and high steady-state accuracy.

[0136] In this embodiment, the target attitude angle of the imaging gimbal is compared with the current actual attitude angle to calculate the required adjustment amount, and then converted into a control command that can be directly driven by the servo motor to gradually adjust the pose of the imaging gimbal.

[0137] The imaging gimbal has two degrees of freedom: pitch and yaw, and its attitude adjustment is controlled by servo motors. Since the imaging target (target brake shoe) dynamically shifts due to train movement, the imaging gimbal must continuously adjust its attitude to maintain optical axis alignment. The most direct and effective method is to read the current attitude angle of the imaging gimbal in real time and compare it with the target attitude angle to obtain the pitch and yaw deviations. These angular deviations are the inputs to the servo control system, which are fed into the built-in PID controller to generate corresponding motor control signals, thereby achieving smooth, high-precision, and fast-response attitude adjustment. The larger the angular deviation, the faster the motor operates; the smaller the angular deviation, the finer the adjustment range. Through this closed-loop adjustment mechanism, the "alignment-compensation-stabilization" process under dynamic conditions is achieved, ensuring that the imaging system is always in the optimal optical alignment state.

[0138] In some embodiments, the spatial pose adjustment command carries the target pitch angle and the target yaw angle. Whether the imaging gimbal has moved to the target pose can be determined in the following way:

[0139] Step 1.1: Obtain the actual pitch angle and actual yaw angle of the imaging gimbal during its movement.

[0140] For example, the actual pitch angle and actual yaw angle of the imaging gimbal are acquired in real time by a high-precision encoder or inertial measurement unit, and the imaging control unit 11 obtains the actual pitch angle and actual yaw angle of the imaging gimbal from the high-precision encoder or inertial measurement unit.

[0141] Step 1.2: Determine the second pitch angle deviation between the actual pitch angle and the target pitch angle; determine the second yaw angle deviation between the actual yaw angle and the target yaw angle.

[0142] For example, let the current sampling time be... Real-time reading of the current encoder feedback angle of the imaging pan-tilt unit Second pitch angle deviation Second pitch angle deviation .

[0143] Step 1.3: If the second pitch angle deviation and the second yaw angle deviation meet the pose positioning conditions, then the imaging gimbal is determined to move to the target pose.

[0144] The conditions for moving to the correct position include that the second pitch angle deviation is less than the pitch angle deviation threshold and the second yaw angle deviation is less than the yaw angle deviation threshold.

[0145] For example, using Indicates the pitch angle deviation threshold, using This represents the yaw angle deviation threshold, which is fixed at 0.005 rad. Setting the threshold too high can affect the alignment, while setting it too low may lead to misjudgment that the imaging gimbal has not moved into position. In practical applications, the threshold can be calibrated based on imaging quality requirements and the performance of the gimbal servo system, and then further fine-tuned based on statistics (such as imaging success rate and false trigger rate).

[0146] exist Less than ,and Less than At that time, the imaging gimbal is moved to the target pose.

[0147] In this embodiment, a closed-loop feedback mechanism is used to determine whether the imaging gimbal has moved to the target pose. This effectively solves the problem in related technologies where the lack of a mechanism to determine the stability of the imaging gimbal pose leads to exposure being triggered before the imaging gimbal has fully moved into position, thus affecting image clarity.

[0148] To more accurately and reliably determine whether the imaging gimbal has moved to the target pose, one implementation further includes: if, within a set duration, both the second pitch angle deviation and the second yaw angle deviation meet the pose movement into position conditions, then the imaging gimbal is determined to have moved to the target pose.

[0149] For example, using This indicates the threshold for the stable hold duration (i.e., the set duration), with a value ranging from 50ms to 100ms. In other words, in... At that time, determine the position and orientation of the imaging gimbal. This indicates the moment when the deviation tolerance window is first entered.

[0150] This implementation method, by setting a duration, requires the angle deviation to remain stable within a tolerance threshold for a certain period before determining that the imaging gimbal is truly in position. This avoids false positioning and misjudgment due to high-precision encoder jitter or critical values, thus preventing false triggering of imaging control. This implementation method ensures the accuracy of imaging timing, guarantees the angle accuracy and clarity of subsequent brake shoe image acquisition, and is a crucial step in achieving closed-loop attitude control.

[0151] In some embodiments, controlling the high-speed imaging device on the imaging pan-tilt unit to image the target brake shoe may include the following steps:

[0152] S2041. Based on the mapping relationship between train speed and exposure time, determine the target exposure time according to the real-time speed.

[0153] The mapping relationship between train speed and exposure time is established using a pre-defined table or function model that establishes the correspondence between train passing speed and optimal exposure time. For example, it is obtained through theoretical calculations combined with on-site calibration experiments, ensuring that clear and motion-free images can be obtained at various speeds.

[0154] For example, the mapping relationship between train speed and exposure time is expressed as follows: , This is the real-time speed, in m / s. The exposure time of the imaging system is expressed in μs. This represents the mapping function obtained through trackside testing, ensuring a balance between image brightness and sharpness at different vehicle speeds.

[0155] It should be noted that the above expression does not represent a physical functional relationship between velocity and time, but rather an empirical mapping table or fitting function obtained through experiments or tests. It is a function that takes the speed value as input and outputs a recommended exposure time. In essence, it is a lookup table process. Although the units on both sides are different, the input and output of the function itself are defined quantities that can be fitted by looking up a table.

[0156] S2042. Generate imaging acquisition instructions based on target exposure time.

[0157] For example, the imaging acquisition command can be presented in the form of a global exposure trigger pulse, which carries the target exposure time.

[0158] S2043. Send an imaging acquisition command to the high-speed imaging device to trigger the high-speed imaging device to image the target brake shoe.

[0159] For example, when a high-speed imaging device receives an imaging acquisition command sent by the imaging control unit, it immediately performs an image acquisition action.

[0160] Understandably, firstly, after detecting that the imaging gimbal has moved into position, the imaging control unit triggers the high-speed imaging device to acquire the target brake shoe image, preventing the target brake shoe from leaving the imaging window due to continued train movement and ensuring that the high-speed imaging device acquires a complete image frame at the most appropriate time. Secondly, to ensure that the image does not exhibit ghosting or excessive darkness due to different train speeds, the imaging control unit searches for a preset mapping relationship between train speed and exposure time based on the current real-time speed, obtains the exposure time matching the current train speed, and sends it to the high-speed imaging device. The faster the train speed, the shorter the exposure time should be to avoid image blurring; when the train speed is slower, the longer the exposure time should be to improve image brightness. Through the coordinated control of these two synchronous actions, not only is the accuracy of the imaging timing guaranteed, but the image quality also remains stable and clear at various operating speeds. It is also a direct response to the aforementioned series of attitude compensation and pose movement positioning judgment operations.

[0161] In this embodiment, after the imaging gimbal is moved into place, precise triggering and adaptive exposure control of image acquisition are achieved to ensure that the imaging timing and imaging parameters are in the optimal state, avoiding overexposure or image ghosting problems, thereby improving the accuracy of the brake shoe status detection results.

[0162] Based on the above embodiments, in one possible implementation, the brake shoe imaging control method further includes:

[0163] S205. Acquire the brake shoe image of the target brake shoe output by the high-speed imaging device.

[0164] For example, the high-speed imaging device synchronously sends the acquired, attitude-compensated target brake shoe image to the imaging control unit.

[0165] S206. Associate the brake shoe image with the real-time speed, target pitch angle, target yaw angle, and imaging timestamp, and output the associated and labeled brake shoe image.

[0166] For example, the imaging control unit performs structured association marking on the brake shoe image obtained in step S205 with the real-time speed, target pitch angle, target yaw angle, and imaging timestamp. The associated and marked brake shoe image will be accompanied by the following metadata information, including: the real-time speed of the current train. Target pitch angle and target yaw angle Imaging timestamp The final generated data structure Represented as:

[0167]

[0168] Considering the need for automatic archiving, analysis, and positioning of each brake shoe image in practical applications, this embodiment of the application binds each brake shoe image with its corresponding key operational status data, including the train's real-time speed at the time of acquisition, attitude compensation angle, and imaging timestamp. This information serves as the metadata of the brake shoe image, forming a structured data frame together with the brake shoe image itself, facilitating subsequent algorithm processing, trackside identification, equipment linkage control, or historical record retrieval.

[0169] Based on the above embodiments, combined with Figure 3 The brake shoe imaging control method provided in the embodiments of this application will be described in detail. Figure 3 This is a schematic flowchart of a brake shoe imaging control method provided in another embodiment of this application, as shown below. Figure 3 As shown, the brake shoe imaging control method includes:

[0170] S301, Obtain the real-time speed of the train.

[0171] For details, please refer to step S201, which will not be repeated here.

[0172] S302. Input the real-time speed into the preset bogie dynamics model, and predict the attitude angle of the bogie corresponding to the target brake shoe at the real-time speed through the bogie dynamics model.

[0173] For details, please refer to step S202, which will not be repeated here.

[0174] S303. Based on the geometric parameters of the target brake shoe's installation position on the corresponding bogie, determine the spatial deflection direction of the normal vector corresponding to the target brake shoe's imaging surface according to the attitude angle.

[0175] S304. Apply the spatial vector inverse solution algorithm to perform spatial coordinate inverse solution on the spatial deflection direction to obtain the target pitch angle and target yaw angle of the imaging gimbal.

[0176] S305. Based on the target pitch angle and target yaw angle, generate spatial pose adjustment commands for the imaging gimbal.

[0177] For example, the initial pitch angle and initial yaw angle of the imaging gimbal are acquired in real time by a high-precision encoder or inertial measurement unit. The imaging control unit 11 obtains the initial pitch angle and initial yaw angle of the imaging gimbal from the high-precision encoder or inertial measurement unit. The first pitch angle deviation of the target pitch angle relative to the initial pitch angle is calculated. Calculate the initial yaw angle Difference with target yaw angle .Will , This is converted into the corresponding angular displacement command for the servo motor. The built-in PID controller generates a sequence of spatial pose adjustment commands to drive the motion of the imaging gimbal's servo motor. This is to control the rotation and adjustment of the imaging gimbal, ensuring that the imaging optical axis is precisely aligned with the normal direction of the target gate tile imaging surface.

[0178] S306. Send the spatial pose adjustment command to the servo driver of the imaging gimbal, and continuously determine whether the imaging gimbal has moved to the target pose during the movement of the imaging gimbal.

[0179] For example, the actual pitch angle and actual yaw angle of the imaging gimbal are obtained during the movement; the second pitch angle deviation between the actual pitch angle and the target pitch angle is determined; the second yaw angle deviation between the actual yaw angle and the target yaw angle is determined; if the second pitch angle deviation is continuously less than the pitch angle deviation threshold and the second yaw angle deviation is continuously less than the yaw angle deviation threshold within a set duration, then the imaging gimbal is determined to have moved to the target pose.

[0180] This step uses the previously generated spatial pose adjustment commands to drive the imaging gimbal servo motor, and a closed-loop feedback mechanism determines whether the target pose has been reached, serving as a prerequisite for image acquisition. During train operation, the imaging gimbal must precisely align with the brake shoe imaging surface after the pose change, and this alignment process relies on the servo system accurately executing the calculated target pitch and yaw angles. Therefore, the imaging control unit sends the spatial pose adjustment command sequence generated in the previous stage to the imaging gimbal servo driver, controlling the servo motor to rotate according to the target pose angle.

[0181] To ensure the reliability of the imaging gimbal's attitude adjustment, a high-precision encoder reads the actual pitch and yaw angles of the imaging gimbal in real time and compares them with the target attitude angles. Only if the angle deviation remains within the set error tolerance range for a continuous period is the gimbal considered truly stable. This process avoids false triggering of imaging due to brief jitter or instability, ensuring the angular accuracy and clarity of subsequent image acquisition. It is a crucial step in achieving closed-loop attitude control.

[0182] When the imaging gimbal is determined to have moved to the target pose, S307 is executed;

[0183] Otherwise, continue executing S305. If, after exceeding the preset time, it is still determined that the imaging gimbal has not moved to the target pose, an exception is thrown, such as outputting an exception message, and relevant personnel are notified to investigate.

[0184] S307. Based on the mapping relationship between train speed and exposure time, determine the target exposure time according to the real-time speed.

[0185] For example, the target exposure time corresponding to the real-time speed can be obtained by looking up the mapping table.

[0186] S308. Generate an imaging acquisition command carrying the target exposure time and send it to the high-speed imaging device to trigger the high-speed imaging device to image the target brake shoe.

[0187] For example, when a high-speed imaging device receives an imaging acquisition command sent by the imaging control unit, it immediately performs an image acquisition action.

[0188] S309: Receive the brake shoe image output by the high-speed imaging device, and then output the brake shoe image after associating and marking it with the real-time speed, target pitch angle, target yaw angle, and imaging timestamp.

[0189] This application embodiment obtains the speed of passing freight cars in real time and combines it with a bogie dynamic model established based on vehicle structural parameters to dynamically predict the pitch and roll angles of the bogie where the target brake shoe is located, thereby achieving advance quantification of attitude changes. Compared with traditional static gimbal control or solutions that rely on image post-processing, this method can actively complete the feedforward correction of the optical axis direction before imaging, effectively avoiding the problem of viewing angle offset caused by changes in vehicle speed or uneven track, and improving the spatial alignment accuracy of the imaging viewpoint.

[0190] Furthermore, an attitude closed-loop control mechanism based on a high-precision encoder is introduced to compare the actual pitch angle and actual yaw angle in real time during the imaging gimbal adjustment process. Dynamic pose positioning conditions are set, and imaging is only triggered after the attitude error has stabilized and converged. This decoupled attitude positioning-imaging control process effectively avoids the problem of starting exposure while the imaging gimbal is still in the dynamic adjustment process, ensuring the clarity and stability of each frame of brake shoe image, making it suitable for high-speed railway freight car scenarios.

[0191] Finally, before image acquisition, the preset mapping relationship between train speed and exposure time is queried based on real-time speed, and the exposure time matching the speed is dynamically set to avoid overexposure or image ghosting. At the same time, after image acquisition, the brake shoe image is synchronously marked with the train's real-time speed, attitude compensation angle, and imaging timestamp to form structured image frame data. This not only improves the image acquisition quality, but also provides accurate and traceable multi-dimensional basic information for subsequent intelligent recognition, trackside monitoring, data management and other systems.

[0192] Next, this application provides a brake shoe imaging control system, still referring to... Figure 1 As shown, the brake shoe imaging control system of this application embodiment includes: an imaging control unit 11, an imaging pan-tilt unit 12, and a high-speed imaging device 13. Wherein:

[0193] Imaging control unit 11 is used to execute the brake shoe imaging control method as described in the above embodiment;

[0194] The imaging gimbal 12 is connected to the imaging control unit and is used to adjust the optical axis direction by adjusting the spatial pose under the control of the imaging control unit in order to compensate for the deflection of the imaging surface caused by the bogie attitude change of the target brake shoe.

[0195] The high-speed imaging device 13 is mounted on the imaging pan-tilt unit and is connected in communication with the imaging control unit. The high-speed imaging device is used to image the target brake shoe under the control of the imaging control unit.

[0196] Figure 4 This is a schematic diagram of the structure of a brake shoe imaging control device provided in an embodiment of this application. The brake shoe imaging control device is applied to the imaging control unit, such as... Figure 4As shown, the brake shoe imaging control device 40 provided in this embodiment includes:

[0197] Module 41 is used to acquire the real-time speed of the train;

[0198] Input module 42 is used to input the real-time speed into the preset bogie dynamics model, and predict the attitude angle of the bogie corresponding to the target brake shoe at the real-time speed through the bogie dynamics model. The bogie dynamics model is constructed based on the secondary suspension system parameters, wheel-rail contact geometry parameters and train dynamics equations of the corresponding train model.

[0199] The generation module 43 is used to generate spatial pose adjustment commands for the imaging gimbal based on the attitude angle. The spatial pose adjustment commands are used to drive the imaging gimbal to adjust the optical axis direction to compensate for the deflection of the imaging surface caused by the bogie attitude change of the target brake shoe.

[0200] The control module 44 is used to control the movement of the imaging gimbal according to the spatial pose adjustment command, and in response to the movement of the imaging gimbal to the target pose, control the high-speed imaging device on the imaging gimbal to image the target brake shoe.

[0201] In one possible implementation, the generation module 43 is specifically used to: determine the spatial deflection direction of the normal vector corresponding to the imaging surface of the target brake shoe based on the geometric parameters of the installation position of the target brake shoe on the corresponding bogie, according to the attitude angle; apply the spatial vector inverse solution algorithm to perform spatial coordinate inverse solution on the spatial deflection direction to obtain the target pitch angle and target yaw angle of the imaging gimbal; and generate a spatial pose adjustment command for the imaging gimbal based on the target pitch angle and target yaw angle.

[0202] In one possible implementation, the generation module 43 is further configured to: acquire the initial pitch angle and initial yaw angle of the imaging gimbal; determine the first pitch angle deviation of the target pitch angle relative to the initial pitch angle, and determine the first yaw angle deviation of the target yaw angle relative to the initial yaw angle; and, based on the built-in PID controller, generate a spatial pose adjustment command for adjusting the spatial pose of the imaging gimbal according to the first pitch angle deviation and the first yaw angle deviation.

[0203] In one possible implementation, the spatial pose adjustment command carries the target pitch angle and the target yaw angle. The imaging gimbal is moved to the target pose by: acquiring the actual pitch angle and the actual yaw angle of the imaging gimbal during the movement; determining the second pitch angle deviation between the actual pitch angle and the target pitch angle; determining the second yaw angle deviation between the actual yaw angle and the target yaw angle; if the second pitch angle deviation and the second yaw angle deviation meet the pose move-to-position conditions, then the imaging gimbal is determined to have moved to the target pose. The pose move-to-position conditions include that the second pitch angle deviation is less than the pitch angle deviation threshold and the second yaw angle deviation is less than the yaw angle deviation threshold.

[0204] In one possible implementation, determining that the imaging gimbal has moved to the target pose includes: if, within a set duration, both the second pitch angle deviation and the second yaw angle deviation satisfy the pose movement into position condition, then the imaging gimbal has moved to the target pose.

[0205] In one possible implementation, the control module 44 is specifically used to: determine the target exposure time based on the real-time speed according to the mapping relationship between train speed and exposure time; generate an imaging acquisition command based on the target exposure time; and send the imaging acquisition command to the high-speed imaging device to trigger the high-speed imaging device to image the target brake shoe.

[0206] See Figure 5 In one possible implementation, the brake shoe imaging control device 40 further includes an output module 45, which is used to acquire the brake shoe image of the target brake shoe output by the high-speed imaging device; associate the brake shoe image with the real-time speed, target pitch angle, target yaw angle and imaging timestamp, and output the associated and marked brake shoe image.

[0207] In one possible implementation, the control module 44 is further configured to: send a spatial pose adjustment command to the servo driver corresponding to the imaging gimbal, so as to drive the servo motor of the imaging gimbal to adjust the pose of the imaging gimbal along the pitch axis and yaw axis.

[0208] The brake shoe imaging control device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0209] Figure 6 This is a schematic diagram of the imaging control unit provided in an embodiment of this application. Figure 6 As shown, the imaging control unit 11 provided in this embodiment includes at least one processor 111 and a memory 112. Optionally, the imaging control unit 11 further includes a communication component 113. The processor 111, the memory 112, and the communication component 113 are connected via a bus 114.

[0210] In a specific implementation, at least one processor 111 executes computer execution instructions stored in memory 112, causing at least one processor 111 to perform the above-described method.

[0211] The specific implementation process of processor 111 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0212] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0213] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0214] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0215] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0216] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0217] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0218] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0219] The division of units is merely a logical functional division; 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 indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0220] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0221] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0222] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0223] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0224] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A brake shoe imaging control method, characterized in that, include: Obtain the train's real-time speed; The real-time speed is input into a preset bogie dynamics model, and the attitude angle of the bogie corresponding to the target brake shoe at the real-time speed is predicted by the bogie dynamics model. The bogie dynamics model is constructed based on the secondary suspension system parameters, wheel-rail contact geometry parameters and train dynamics equations of the train model corresponding to the train type. Based on the attitude angle, a spatial pose adjustment command for the imaging gimbal is generated. The spatial pose adjustment command is used to drive the imaging gimbal to adjust the optical axis direction in order to compensate for the image plane deflection of the target brake shoe caused by the bogie attitude change. According to the spatial pose adjustment command, the imaging gimbal is controlled to move, and in response to the imaging gimbal moving to the target pose, the high-speed imaging device on the imaging gimbal is controlled to image the target brake shoe.

2. The brake shoe imaging control method according to claim 1, characterized in that, The step of generating spatial pose adjustment commands for the imaging gimbal based on the attitude angle includes: Based on the geometric parameters of the installation position of the target brake shoe on the corresponding bogie, the spatial deflection direction of the normal vector corresponding to the imaging plane of the target brake shoe is determined according to the attitude angle. By applying the spatial vector inverse algorithm, the spatial coordinate inverse solution is performed on the spatial deflection direction to obtain the target pitch angle and target yaw angle of the imaging gimbal. Based on the target pitch angle and target yaw angle, a spatial pose adjustment command for the imaging gimbal is generated.

3. The brake shoe imaging control method according to claim 2, characterized in that, The step of generating spatial pose adjustment commands for the imaging gimbal based on the target pitch angle and target yaw angle includes: Obtain the initial pitch angle and initial yaw angle of the imaging gimbal; Determine the first pitch angle deviation of the target pitch angle relative to the initial pitch angle, and determine the first yaw angle deviation of the target yaw angle relative to the initial yaw angle; Based on the built-in PID controller, a spatial pose adjustment command is generated to adjust the spatial pose of the imaging gimbal according to the first pitch angle deviation and the first yaw angle deviation.

4. The brake shoe imaging control method according to any one of claims 1 to 3, characterized in that, The spatial pose adjustment command carries the target pitch angle and the target yaw angle, and the imaging gimbal is moved to the target pose in the following way: Obtain the actual pitch angle and actual yaw angle of the imaging gimbal during its movement; Determine the second pitch angle deviation between the actual pitch angle and the target pitch angle; Determine the second yaw angle deviation between the actual yaw angle and the target yaw angle; If the second pitch angle deviation and the second yaw angle deviation meet the pose shifting conditions, then the imaging gimbal is determined to have moved to the target pose. The pose shifting conditions include the second pitch angle deviation being less than the pitch angle deviation threshold and the second yaw angle deviation being less than the yaw angle deviation threshold.

5. The brake shoe imaging control method according to claim 4, characterized in that, Determining the motion of the imaging gimbal to the target pose includes: If, within a set duration, both the second pitch angle deviation and the second yaw angle deviation satisfy the pose positioning condition, then the imaging gimbal is determined to have moved to the target pose.

6. The brake shoe imaging control method according to any one of claims 1 to 3, characterized in that, The method of controlling the high-speed imaging device on the imaging pan-tilt unit to image the target brake shoe includes: Based on the mapping relationship between train speed and exposure time, the target exposure time is determined according to the real-time speed; Based on the target exposure time, an imaging acquisition command is generated; Send the imaging acquisition command to the high-speed imaging device to trigger the high-speed imaging device to image the target brake shoe.

7. The brake shoe imaging control method according to any one of claims 1 to 3, characterized in that, Also includes: Acquire the brake shoe image of the target brake shoe output by the high-speed imaging device; The brake shoe image is associated with the real-time speed, the target pitch angle, the target yaw angle, and the imaging timestamp, and the associated and labeled brake shoe image is output.

8. The brake shoe imaging control method according to any one of claims 1 to 3, characterized in that, The step of controlling the movement of the imaging gimbal according to the spatial pose adjustment command includes: The spatial pose adjustment command is sent to the servo driver corresponding to the imaging gimbal to drive the servo motor of the imaging gimbal to adjust the pose of the imaging gimbal along the pitch axis and yaw axis.

9. A brake shoe imaging control system, characterized in that, include: An imaging control unit is used to perform the brake shoe imaging control method as described in any one of claims 1 to 8 above; An imaging gimbal, which is communicatively connected to the imaging control unit, is used to adjust the optical axis direction by adjusting the spatial pose under the control of the imaging control unit in order to compensate for the deflection of the imaging surface caused by the bogie attitude change of the target brake shoe. A high-speed imaging device is mounted on the imaging pan-tilt unit and is communicatively connected to the imaging control unit. The high-speed imaging device is used to image the target brake shoe under the control of the imaging control unit.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed, are used to implement the brake shoe imaging control method as described in any one of claims 1 to 8.

Images