Monitoring method, system, equipment, medium and program for wheel-rail contact track
Through multi-computer point cloud acquisition and feature matching technology, combined with profile curve fitting and geometric constraint algorithm, the problems of low monitoring efficiency of wheel and rail contact trajectory and inaccurate contact points in the existing technology are solved, and high-precision contact point positioning and monitoring efficiency are improved.
Patent Information
- Application Number
- CN202411651170.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-19
- Publication Date
- 2025-02-25
AI Technical Summary
The existing railway wheel-rail contact trajectory monitoring technology has problems such as high installation costs, difficulty in data calibration, low reliability and inaccurate contact points, resulting in low monitoring efficiency.
The point cloud data of wheel and rail contacts are obtained by multi-machine point cloud acquisition method, point cloud classification and coordinate system consistency are performed through feature matching, dimensionality reduction is used for contour curve fitting, and high-precision contact point calculation is performed through geometric constraint algorithm.
The monitoring efficiency of wheel and rail contact trajectory is improved, high-precision contact point positioning is achieved, monitoring costs are reduced and system reliability is improved.
Smart Images
Figure CN119509408A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of photoelectric detection technology, and in particular to a method, system, device, medium and program for monitoring a wheel-rail contact trajectory. Background Art
[0002] As an important national infrastructure and livelihood project, railways have the advantages of large transportation volume, fast speed, safety and reliability compared to other modes of transportation. They play an important role in medium and long distance passenger transportation and long distance and large-volume freight transportation.
[0003] The heavy load and axle weight of heavy-duty railways directly lead to the intensification of the dynamic interaction between wheels and rails, resulting in damage to the vehicle tread, increased wheel wear and track wear, which directly affects the stability, safety and operation and maintenance costs of freight cars. In recent years, computer vision technology has developed rapidly, and the use of image detection technology can effectively detect the wheel-rail contact relationship online. The wheel-rail contact trajectory, as the most direct reflection of the wheel-rail contact relationship, plays a key role in the in-depth analysis of the dynamic behavior and safety of wheelset operation. Therefore, it is particularly important to study the continuous and online detection method of wheel-rail contact points based on image detection technology to monitor the wheel-rail contact trajectory online.
[0004] Traditional railway wheel-rail contact trajectory monitoring technologies are mostly sensor-based monitoring methods or image-based monitoring methods. In practical applications, sensor-based monitoring methods have high installation costs, difficult data calibration, and low reliability. Image-based monitoring methods may cause inaccurate contact point positions due to contour deformation of the rail and wheel tread, which may lead to low efficiency in monitoring the wheel-rail contact trajectory. Summary of the invention
[0005] The present invention provides a monitoring method, system, device, medium and program for wheel-rail contact trajectory, which obtains point cloud data containing detailed information of wheel-rail contact by a multi-machine point cloud acquisition method, realizes point cloud classification by feature matching, realizes coordinate system of point cloud data by point cloud matching, realizes dimensionality reduction of point cloud data by profile curve fitting, and realizes high-precision contact point calculation by geometric constraint algorithm, thereby improving the monitoring efficiency of wheel-rail contact trajectory.
[0006] In a first aspect, the present disclosure provides a method for monitoring a wheel-rail contact trajectory, comprising: performing multi-machine point cloud measurement on a wheel-rail using a preset point cloud sensor to obtain a wheel profile point cloud set, a rail profile point cloud set and a wheel-rail profile point cloud set; splitting the wheel-rail profile point cloud set into a wheel side point cloud set and a rail side point cloud set; performing point cloud matching and coordinate transformation on the wheel profile point cloud set using the wheel side point cloud set to obtain a standard wheel point cloud set, performing point cloud matching and coordinate transformation on the rail profile point cloud set using the rail side point cloud set to obtain a standard rail point cloud set; fitting the standard wheel point cloud set into a vehicle profile curve set, and fitting the standard rail point cloud set into a rail profile curve set; calculating a rail contact point set between the vehicle profile curve set and the rail profile curve set, and generating a wheel-rail contact trajectory according to the rail contact point set.
[0007] In some embodiments, splitting the wheel-rail contour point cloud set into a wheel side point cloud set and a rail side point cloud set includes: extracting a wheel-rail point cloud feature set from the wheel-rail contour point cloud set; performing feature classification on the wheel-rail point cloud feature set to obtain a wheel point cloud feature set and a rail point cloud feature set; using part of the point clouds corresponding to the wheel point cloud feature set in the wheel-rail contour point cloud set as the wheel side point cloud set; using part of the point clouds corresponding to the rail point cloud feature set in the wheel-rail contour point cloud set as the rail side point cloud set.
[0008] In some embodiments, the use of the wheel side point cloud set to perform point cloud matching and coordinate transformation on the wheel contour point cloud set to obtain a standard wheel point cloud set includes: performing point cloud denoising and point cloud smoothing on the wheel side point cloud set and the wheel contour point cloud set, respectively, to obtain a denoised wheel side point cloud set and a denoised wheel contour point cloud set; performing feature point extraction and feature point encoding on the denoised wheel side point cloud set and the denoised wheel contour point cloud set, respectively, to obtain a wheel side feature point group set and a wheel contour feature point group set; performing point cloud matching on the wheel contour feature point group set using the wheel side feature point group set to obtain a wheel rotation matrix set and a wheel translation matrix set; and performing coordinate transformation on the wheel contour point cloud set using the wheel rotation matrix set and the wheel translation matrix set to obtain a standard wheel point cloud set.
[0009] In some embodiments, the use of the wheel side feature point set to perform point cloud matching on the wheel contour feature point set to obtain a wheel rotation matrix set and a wheel translation matrix set includes: selecting the wheel side feature point sets in the wheel side feature point set one by one as the target side feature point set, and taking the wheel contour feature point group corresponding to the target side feature point set in the wheel contour feature point set as the target contour feature point group; performing feature point registration on the target contour feature point group using the target side feature point group to obtain a registered wheel side point group; and selecting the wheel side feature point group from the target side feature point group. A registered wheel contour point group corresponding to the registered wheel side point group is screened out from the registered wheel contour feature point group; a point cloud matching loss model is generated according to the registered wheel side point group and the registered wheel contour point group; a wheel rotation matrix and a wheel translation matrix corresponding to the registered wheel contour point group are calculated according to the point cloud matching loss model; the wheel rotation matrices corresponding to all target side feature point groups in the wheel side feature point group set are collected into a wheel rotation matrix set, and the wheel translation matrices corresponding to all target side feature point groups in the wheel side feature point group set are collected into a wheel translation matrix set.
[0010] In some embodiments, fitting the standard wheel point cloud set into a vehicle contour curve set includes: selecting standard wheel point clouds in the standard wheel point cloud set one by one as target standard wheel point clouds, and calculating the wheel pair axis of the target standard wheel point cloud; generating a wheel vertical plane of the target standard wheel point cloud according to the wheel pair axis; gathering the intersection points of the target standard wheel point cloud and the wheel vertical plane into a wheel fitting point cloud group; sorting the wheel fitting point cloud group in ascending order of the horizontal coordinate to obtain a wheel fitting point cloud sequence; performing curve fitting on the wheel fitting point cloud sequence to obtain a vehicle contour curve; and gathering the vehicle contour curves corresponding to each target standard wheel point cloud in the standard wheel point cloud set into a vehicle contour curve set.
[0011] In some embodiments, the step of calculating the set of contact points between the vehicle profile curve set and the rail profile curve set comprises: selecting vehicle profile curves in the vehicle profile curve set one by one as target vehicle profile curves, and taking the rail profile curve corresponding to the target vehicle profile curve in the rail profile curve set as the target rail profile curve; calculating the vertical distance between the target vehicle profile curve and the target rail profile curve; calculating the extreme value of the vertical distance to obtain a vertical distance extreme value group; and calculating the vertical distance extreme value from the vertical distance extreme value. The minimum vertical distance extreme value is screened out from the group, and the difference calculation is performed using the minimum vertical distance extreme value and the remaining vertical distance extreme values in the vertical distance extreme value group to obtain a distance difference value group; it is determined whether there is a distance difference value less than a preset wheel-rail contact penetration amount in the distance difference value group; if not, a rail contact point is generated according to the minimum vertical distance extreme value; if so, the distance differences in the distance difference value group that are less than the wheel-rail contact penetration amount are aggregated into a standard distance difference value group, and a rail contact point is generated according to the standard distance difference value group and the minimum vertical distance extreme value.
[0012] In some embodiments, fitting the standard rail point cloud set into a rail profile curve set includes: selecting standard rail point clouds in the standard rail point cloud set one by one as target standard rail point clouds, and calculating the rail centerline of the target standard rail point cloud; generating a rail vertical plane of the target standard rail point cloud according to the rail centerline; collecting the intersection points of the target standard rail point cloud and the rail vertical plane into a rail fitting point cloud group; sorting the rail fitting point cloud group in ascending order of the horizontal coordinate to obtain a rail fitting point cloud sequence; performing curve fitting on the rail fitting point cloud sequence to obtain a rail profile curve; and collecting the rail profile curves corresponding to each target standard rail point cloud in the standard rail point cloud set into a rail profile curve set.
[0013] In a second aspect, the present disclosure provides a monitoring system for wheel-rail contact trajectory, including: a point cloud measurement module, used to perform multi-machine point cloud measurement on the wheel and rail using a preset point cloud sensor, and obtain a wheel profile point cloud set, a rail profile point cloud set and a wheel-rail profile point cloud set; a point cloud segmentation module, used to split the wheel-rail profile point cloud set into a wheel side point cloud set and a rail side point cloud set; a point cloud matching module, used to perform point cloud matching and coordinate transformation on the wheel profile point cloud set using the wheel side point cloud set, and obtain a standard wheel point cloud set, and perform point cloud matching and coordinate transformation on the rail profile point cloud set using the rail side point cloud set, and obtain a standard rail point cloud set; a curve fitting module, used to fit the standard wheel point cloud set into a vehicle profile curve set, and fit the standard rail point cloud set into a rail profile curve set; a trajectory calculation module, used to calculate a rail contact point set between the vehicle profile curve set and the rail profile curve set, and generate a wheel-rail contact trajectory according to the rail contact point set.
[0014] In a third aspect, the present disclosure provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of a method for monitoring wheel-rail contact trajectory described in the above aspect.
[0015] In a fourth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for monitoring a wheel-rail contact trajectory as described in the above aspect.
[0016] In a fifth aspect, the present disclosure provides a computer program product, including a computer program / instructions, which, when executed by a processor, implements the steps of a method for monitoring a wheel-rail contact trajectory described in the above aspects.
[0017] The present disclosure provides a wheel-rail contact trajectory monitoring method, system, device, medium and program, which can extract the quality of the finished product after the printer prints different 3D models according to different printing paths in different printing environments by performing multi-dimensional type feature extraction and model quality evaluation, thereby facilitating subsequent model training for solving high-quality printing path models.
[0018] By acquiring the wheel contour point cloud set, the rail contour point cloud set and the wheel-rail contour point cloud set, the point cloud data can be used to realize multi-angle data collection of the wheel and the rail, and contain the corresponding position information, thereby improving the accuracy of contact trajectory monitoring. By splitting the wheel-rail contour point cloud set into the wheel side point cloud set and the rail side point cloud set, the wheel-rail contour point cloud set can be split, which facilitates the subsequent acquisition of complete railway wheel point cloud and railway rail point cloud containing multiple perspectives, thereby realizing accurate contact point matching.
[0019] By performing point cloud matching on the wheel profile point cloud set and converting the coordinates into a standard wheel point cloud set and performing point cloud matching on the rail profile point cloud set and converting the coordinates into a standard rail point cloud set, the wheel profile point cloud set and the rail profile point cloud set can be unified into the same reference coordinate system, thereby facilitating the subsequent calculation of contact points.
[0020] By fitting the standard wheel point cloud set into a vehicle profile curve set, and fitting the standard rail point cloud set into a rail profile curve set, dimensionality reduction processing of point cloud data is achieved, thereby reducing the complexity of contact point calculation and improving the efficiency of contact trajectory calculation. By calculating the rail contact point set between the vehicle profile curve set and the rail profile curve set, accurate contact point positioning can be achieved, thereby improving the efficiency of wheel-rail contact trajectory monitoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The present disclosure will be described in more detail below based on embodiments and with reference to the accompanying drawings:
[0022] Figure 1 A schematic flow chart of a method for monitoring wheel-rail contact trajectory according to a first embodiment of the present disclosure is shown.
[0023] Figure 2 A schematic diagram of the process of training a printing path planning model in the first embodiment of the present disclosure is shown.
[0024] Figure 3 A schematic diagram of the process of performing printing simulation and real-time process rendering in the first embodiment of the present disclosure is shown.
[0025] Figure 4 A schematic flow chart of a method for monitoring wheel-rail contact trajectory according to a second embodiment of the present disclosure is shown.
[0026] Figure 5 A functional module diagram of a wheel-rail contact trajectory monitoring system according to a third embodiment of the present disclosure is shown.
[0027] In the drawings, the same reference numerals are used for the same components, and the drawings are not drawn to scale. DETAILED DESCRIPTION
[0028] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, and to fully understand and implement how the present disclosure applies technical means to solve technical problems and achieve the corresponding technical effects, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only embodiments of a part of the present disclosure, not all of the embodiments. The embodiments of the present disclosure and the various features in the embodiments can be combined with each other without conflict, and the technical solutions formed are all within the scope of protection of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present disclosure.
[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.
[0030] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0031] Embodiment 1
[0032] Figure 1 A schematic flow chart of a method for monitoring wheel-rail contact trajectory provided in an embodiment of the present disclosure.
[0033] like Figure 1 As shown, a method for monitoring wheel-rail contact trajectory comprises:
[0034] S1. Use a preset point cloud sensor to perform multi-machine point cloud measurement on the wheel-rail to obtain a wheel profile point cloud set, a rail profile point cloud set, and a wheel-rail profile point cloud set.
[0035] In detail, the point cloud sensor is a point cloud device used to measure the contact trajectory between the wheel and the rail. The point cloud sensor can be a laser scanner, a structured light scanner or a time-of-flight camera. The point cloud sensor can be composed of three components, namely, a point cloud sensor for the wheel, a point cloud sensor for the rail, and a point cloud sensor for the relative position of the wheel and the rail. The point cloud sensors are respectively installed on both sides of the wheel pair being measured.
[0036] Specifically, each wheel contour point cloud in the wheel contour point cloud set is point cloud data obtained by a point cloud sensor installed on the vehicle body collecting data on the tread of the wheel at a certain moment, each rail contour point cloud in the rail contour point cloud set is point cloud data obtained by a point cloud sensor installed on the vehicle body collecting data on the rail surface at a certain moment, and each wheel-rail contour point cloud in the wheel-rail contour point cloud set is point cloud data obtained by a point cloud sensor installed on the vehicle body collecting data on the wheel and the rail at a certain moment.
[0037] For details, refer to Figure 2 , which is a schematic diagram of the operation of the point cloud sensor performing multi-machine point cloud measurement on the wheel-rail, wherein t refers to the point cloud sensor, L refers to the wheel, and G refers to the rail.
[0038] In the embodiment of the present invention, by acquiring the wheel contour point cloud set, the rail contour point cloud set and the wheel-rail contour point cloud set, point cloud data can be used to realize multi-angle data collection of wheels and rails, and include corresponding position information, thereby improving the accuracy of contact trajectory monitoring.
[0039] S2. Splitting the wheel-rail contour point cloud set into a wheel side point cloud set and a rail side point cloud set.
[0040] In detail, each wheel side point cloud in the wheel side point cloud set corresponds to the point cloud of the wheel part of each wheel-rail contour point cloud in the wheel-rail contour point cloud set, and each rail side point cloud in the rail side point cloud set corresponds to the point cloud of the rail part of each wheel-rail contour point cloud in the wheel-rail contour point cloud set.
[0041] In an embodiment of the present invention, splitting the wheel-rail contour point cloud set into a wheel side point cloud set and a rail side point cloud set includes: extracting a wheel-rail point cloud feature set from the wheel-rail contour point cloud set; performing feature classification on the wheel-rail point cloud feature set to obtain a wheel point cloud feature set and a rail point cloud feature set; using a portion of point clouds corresponding to the wheel point cloud feature set in the wheel-rail contour point cloud set as a wheel side point cloud set; and using a portion of point clouds corresponding to the rail point cloud feature set in the wheel-rail contour point cloud set as a rail side point cloud set.
[0042] Specifically, each wheel-rail point cloud feature in the wheel-rail point cloud feature set corresponds to the curvature feature, signal feature and point cloud feature of each wheel-rail contour point cloud in the wheel-rail contour point cloud set.
[0043] In detail, feature classification can be performed using a feature mapping classification method based on a K-nearest neighbor algorithm or a trained support vector machine model (SVM), wherein the wheel point cloud feature set is a feature set in the wheel-rail point cloud feature set that satisfies the characteristics of the railway wheel part, and the rail point cloud feature set is a feature set in the wheel-rail point cloud feature set that satisfies the characteristics of the railway rail part.
[0044] In the embodiment of the present invention, by splitting the wheel-rail contour point cloud set into a wheel side point cloud set and a rail side point cloud set, the wheel-rail contour point cloud set can be split, which facilitates the subsequent acquisition of a complete railway wheel point cloud and railway rail point cloud containing multiple perspectives, thereby achieving accurate contact point matching.
[0045] S3. Use the wheel side point cloud set to perform point cloud matching and coordinate transformation on the wheel contour point cloud set to obtain a standard wheel point cloud set, and use the rail side point cloud set to perform point cloud matching and coordinate transformation on the rail contour point cloud set to obtain a standard rail point cloud set.
[0046] In detail, the standard wheel point cloud set is a wheel contour point cloud set with the coordinate system of the wheel side point cloud set as the reference coordinate system, and the standard rail point cloud set is a rail contour point cloud set with the coordinate system of the rail side point cloud set as the reference coordinate system.
[0047] In an embodiment of the present invention, the use of the wheel side point cloud set to perform point cloud matching and coordinate transformation on the wheel contour point cloud set to obtain a standard wheel point cloud set includes: performing point cloud denoising and point cloud smoothing on the wheel side point cloud set and the wheel contour point cloud set, respectively, to obtain a denoised wheel side point cloud set and a denoised wheel contour point cloud set; performing feature point extraction and feature point encoding on the denoised wheel side point cloud set and the denoised wheel contour point cloud set, respectively, to obtain a wheel side feature point group set and a wheel contour feature point group set; performing point cloud matching on the wheel contour feature point group set using the wheel side feature point group set to obtain a wheel rotation matrix set and a wheel translation matrix set; and performing coordinate transformation on the wheel contour point cloud set using the wheel rotation matrix set and the wheel translation matrix set to obtain a standard wheel point cloud set.
[0048] Specifically, statistical filtering or radius filtering can be used to denoise the point cloud, moving least squares or Gaussian filtering can be used to smooth the point cloud, scale-invariant feature transform (SIFT) or Harris corner detection algorithm can be used to extract feature points, and fast point feature histograms (FPFH) or signature of histograms (SHOT) can be used to encode feature points.
[0049] In detail, each wheel side feature point group in the wheel side feature point group set corresponds to a set of point cloud feature points of a denoised wheel side point cloud in the denoised wheel side point cloud set, the point cloud feature point refers to a feature point in the point cloud that can describe the point cloud features, and the use of the wheel rotation matrix set and the wheel translation matrix set to perform coordinate transformation on the wheel contour point cloud set to obtain a standard wheel point cloud set refers to performing a translation transformation on the wheel contour point cloud set using the wheel translation matrix set, and performing a rotation transformation on the wheel contour point cloud set after the translation transformation using the wheel rotation matrix set to obtain a standard wheel point cloud set.
[0050] Specifically, the use of the wheel side feature point set to perform point cloud matching on the wheel contour feature point set to obtain a wheel rotation matrix set and a wheel translation matrix set includes: selecting the wheel side feature point sets in the wheel side feature point set one by one as the target side feature point set, and taking the wheel contour feature point group corresponding to the target side feature point set in the wheel contour feature point set as the target contour feature point group; using the target side feature point group to perform feature point registration on the target contour feature point group to obtain a registered wheel side point group; A registered wheel contour point group corresponding to the registered wheel side point group is screened out from the contour feature point group; a point cloud matching loss model is generated according to the registered wheel side point group and the registered wheel contour point group; a wheel rotation matrix and a wheel translation matrix corresponding to the registered wheel contour point group are calculated according to the point cloud matching loss model; the wheel rotation matrices corresponding to all target side feature point groups in the wheel side feature point group set are collected into a wheel rotation matrix set, and the wheel translation matrices corresponding to all target side feature point groups in the wheel side feature point group set are collected into a wheel translation matrix set.
[0051] In detail, the feature point registration refers to extracting feature points from the target side feature point group that match the target contour feature point group.
[0052] Specifically, the point cloud matching loss model is as follows:
[0053]
[0054] Wherein, Loss is the point cloud matching loss corresponding to the point cloud matching loss model, n is the total number of point clouds in the registered wheel side point group, i is the point cloud index, |||| is the norm symbol, q i is the i-th point cloud in the registered wheel side point group, R is the wheel rotation matrix, p i is the i-th point cloud in the registered wheel contour point group, and T refers to the wheel translation matrix.
[0055] In detail, the point cloud matching loss model can be decomposed using the singular value decomposition (SVD) method to obtain the wheel rotation matrix and the wheel translation matrix. The wheel rotation matrix and the wheel translation matrix are basic tools in geometric transformation for describing the position and direction changes of objects in space, that is, geometric transformation tools for describing the relative position of the wheel side point cloud set and the wheel contour point cloud set.
[0056] Specifically, the method of using the rail side point cloud set to perform point cloud matching and coordinate transformation on the rail contour point cloud set to obtain a standard rail point cloud set is consistent with the method of using the wheel side point cloud set to perform point cloud matching and coordinate transformation on the wheel contour point cloud set to obtain a standard wheel point cloud set, which will not be repeated here.
[0057] In an embodiment of the present invention, by performing point cloud matching on the wheel contour point cloud set and converting coordinates into a standard wheel point cloud set, and by performing point cloud matching on the rail contour point cloud set and converting coordinates into a standard rail point cloud set, the wheel contour point cloud set and the rail contour point cloud set can be unified into the same reference coordinate system, thereby facilitating subsequent calculation of contact points.
[0058] S4. Fitting the standard wheel point cloud set into a wheel profile curve set, and fitting the standard rail point cloud set into a rail profile curve set.
[0059] Specifically, each vehicle profile curve in the vehicle profile curve set is an intersection line formed by cutting the railway wheel with a vertical plane through the axis of the railway wheel and the cutting surface, and each rail profile curve in the rail profile curve set is an intersection line obtained by cutting the rail with a plane perpendicular to the center line of the rail.
[0060] In detail, the step of fitting the standard wheel point cloud set into a vehicle contour curve set includes: selecting the standard wheel point clouds in the standard wheel point cloud set one by one as target standard wheel point clouds, and calculating the wheel pair axis of the target standard wheel point cloud; generating the wheel vertical plane of the target standard wheel point cloud according to the wheel pair axis; gathering the intersection points of the target standard wheel point cloud and the wheel vertical plane into a wheel fitting point cloud group; sorting the wheel fitting point cloud group in ascending order of the horizontal coordinate to obtain a wheel fitting point cloud sequence; performing curve fitting on the wheel fitting point cloud sequence to obtain a vehicle contour curve; and gathering the vehicle contour curves corresponding to each target standard wheel point cloud in the standard wheel point cloud set into a vehicle contour curve set.
[0061] Specifically, the calculation of the wheel pair axis includes performing point cloud centering on the target standard wheel point cloud to obtain a center wheel point cloud; performing covariance operation on the center wheel point cloud to obtain a wheel point cloud covariance matrix; performing eigenvalue decomposition on the wheel point cloud covariance to obtain wheel point cloud eigenvalues and wheel point cloud eigenvectors; calculating the wheel pair axis based on the wheel point cloud eigenvalues and the wheel point cloud eigenvectors, and the wheel vertical plane refers to a plane perpendicular to the wheel plane and passing through the wheel pair axis.
[0062] In detail, the curve fitting method may be a fitting method based on a cubic spline function, including performing curve fitting on the wheel fitting point cloud sequence using the following cubic spline fitting function to obtain wheel fitting parameters, and generating a wheel profile curve according to the wheel fitting parameters:
[0063]
[0064] Among them, x k+1 is the horizontal coordinate of the k+1th point cloud in the wheel fitting point cloud sequence, x k is the horizontal coordinate of the kth point cloud in the wheel fitting point cloud sequence, x k+2 is the horizontal coordinate of the k+2th point cloud in the wheel fitting point cloud sequence, M k-1 、M k and M k+1 is the wheel fitting parameter, y k+1 is the ordinate of the k+1th point cloud in the wheel fitting point cloud sequence, y k is the ordinate of the kth point cloud in the wheel fitting point cloud sequence, y k-1 is the ordinate of the k-1th point cloud in the wheel fitting point cloud sequence.
[0065] Specifically, the step of fitting the standard rail point cloud set into a rail profile curve set comprises: selecting the standard rail point clouds in the standard rail point cloud set as target standard rail point clouds one by one, and calculating the rail centerline of the target standard rail point cloud; generating the rail vertical plane of the target standard rail point cloud according to the rail centerline; collecting the intersection points of the target standard rail point cloud and the rail vertical plane into a rail fitting point cloud group; sorting the rail fitting point cloud group in ascending order of the horizontal coordinate to obtain a rail fitting point cloud sequence; performing curve fitting on the rail fitting point cloud sequence to obtain a rail profile curve; and collecting the rail profile curves corresponding to each target standard rail point cloud in the standard rail point cloud set into a rail profile curve set.
[0066] In the embodiment of the present invention, by fitting the standard wheel point cloud set into a wheel profile curve set, and fitting the standard rail point cloud set into a rail profile curve set, dimensionality reduction processing of point cloud data is achieved, thereby reducing the complexity of contact point calculation and improving the efficiency of contact trajectory calculation.
[0067] S5. Calculate a wheel-rail contact point set between the vehicle profile curve set and the rail profile curve set, and generate a wheel-rail contact trajectory according to the wheel-rail contact point set.
[0068] In detail, each trolley contact point in the trolley contact point set corresponds to a contact point between each vehicle profile curve in the vehicle profile curve set and each rail profile curve in the rail profile curve set.
[0069] Reference Figure 4 As shown, it is a schematic diagram of the contact point of the rail, wherein a refers to the vehicle profile curve in the vehicle profile curve set, b refers to the rail profile curve in the rail profile curve set, c refers to the interface of the wheel-rail profile point cloud in the wheel-rail profile point cloud set, and d refers to the contact point of the rail.
[0070] Specifically, the step of calculating the contact point set between the vehicle profile curve set and the rail profile curve set comprises: selecting vehicle profile curves in the vehicle profile curve set one by one as target vehicle profile curves, and taking the rail profile curve corresponding to the target vehicle profile curve in the rail profile curve set as the target rail profile curve; calculating the vertical distance between the target vehicle profile curve and the target rail profile curve; performing extreme value calculation on the vertical distance to obtain a vertical distance extreme value group; and selecting the target rail profile curve from the vertical distance extreme value group. The minimum vertical distance extreme value is screened out, and the difference calculation is performed using the minimum vertical distance extreme value and the remaining vertical distance extreme values in the vertical distance extreme value group to obtain a distance difference value group; it is determined whether there is a distance difference value less than a preset wheel-rail contact penetration amount in the distance difference value group; if not, a rail contact point is generated according to the minimum vertical distance extreme value; if so, the distance differences in the distance difference value group that are less than the wheel-rail contact penetration amount are aggregated into a standard distance difference value group, and a rail contact point is generated according to the standard distance difference value group and the minimum vertical distance extreme value.
[0071] In detail, the wheel-rail contact penetration amount is a preset distance threshold for judging whether the wheel-rail is in contact, generating the rail contact point according to the minimum vertical distance extreme value refers to using the horizontal coordinate and the vertical coordinate corresponding to the minimum vertical distance extreme value as the coordinates of the rail contact point, and generating the rail contact point according to the standard distance difference group and the minimum vertical distance extreme value refers to using the coordinates of the minimum vertical distance extreme value and each distance difference in the standard distance difference group as the coordinates of the corresponding rail contact point.
[0072] Specifically, generating the wheel-rail contact trajectory according to the rail contact point set means obtaining a primary contact trajectory after sorting the rail contact points in the rail contact point set in chronological order, and performing Kalman filtering on the primary contact trajectory to obtain the wheel-rail contact trajectory.
[0073] In the embodiment of the present invention, by calculating the set of contact points between the vehicle profile curve set and the rail profile curve set, accurate contact point positioning can be achieved, thereby improving the efficiency of wheel-rail contact trajectory monitoring.
[0074] Embodiment 2
[0075] On the basis of the above embodiments, in order to more clearly understand the present disclosure, refer to Figure 4 As shown, the following further explains the situation in which the first embodiment of the present disclosure needs to monitor the contact trajectory more quickly through a second embodiment.
[0076] S41, using a preset point cloud sensor to perform multi-machine point cloud measurement on the wheel-rail to obtain a wheel profile point cloud set, a rail profile point cloud set and a wheel-rail profile point cloud set.
[0077] S42, splitting the wheel-rail contour point cloud set into a wheel side point cloud set and a rail side point cloud set.
[0078] S43, using the wheel side point cloud set to perform grid matching and coordinate transformation on the wheel contour point cloud set to obtain a standard wheel point cloud set, and using the rail side point cloud set to perform grid matching and coordinate transformation on the rail contour point cloud set to obtain a standard rail point cloud set.
[0079] Specifically, the method of using the wheel side point cloud set to perform grid matching and coordinate transformation on the wheel contour point cloud set to obtain a standard wheel point cloud set includes: performing point cloud denoising and point cloud smoothing on the wheel side point cloud set to obtain a denoised wheel side point cloud set; performing point cloud denoising and point cloud smoothing on the wheel contour point cloud set to obtain a denoised wheel contour point cloud set; performing grid division and grid distribution feature extraction on the denoised wheel side point cloud set to obtain a wheel side grid feature set; performing grid division and grid distribution feature extraction on the denoised wheel contour point cloud set to obtain a wheel contour grid feature set; performing point cloud matching on the wheel contour grid feature set using the wheel side grid feature set to obtain a wheel rotation matrix set and a wheel translation matrix set; performing coordinate transformation on the wheel contour point cloud set using the wheel rotation matrix set and the wheel translation matrix set to obtain a standard wheel point cloud set.
[0080] In detail, the grid division refers to dividing the point cloud into grid units of fixed size, the method of extracting the grid distribution features can be a normal distribution calculation method based on the mean or a normal distribution calculation method based on the covariance, and the point cloud matching method can be a normal distribution transformation (Normal Distributions Transform, referred to as NDT) method.
[0081] Specifically, the method of using the rail side point cloud set to perform grid matching and coordinate transformation on the rail contour point cloud set to obtain a standard rail point cloud set includes: performing point cloud denoising and point cloud smoothing on the rail side point cloud set to obtain a denoised rail side point cloud set; performing point cloud denoising and point cloud smoothing on the rail contour point cloud set to obtain a denoised rail contour point cloud set; performing grid division and grid distribution feature extraction on the denoised rail side point cloud set to obtain a rail side grid feature set; performing grid division and grid distribution feature extraction on the denoised rail contour point cloud set to obtain a rail contour grid feature set; performing point cloud matching on the rail contour grid feature set using the rail side grid feature set to obtain a rail rotation matrix set and a rail translation matrix set; performing coordinate transformation on the rail contour point cloud set using the rail rotation matrix set and the rail translation matrix set to obtain a standard rail point cloud set.
[0082] S44, fitting the standard wheel point cloud set into a wheel profile curve set, and fitting the standard rail point cloud set into a rail profile curve set.
[0083] S45, calculating a wheel-rail contact point set between the vehicle profile curve set and the rail profile curve set, and generating a wheel-rail contact trajectory according to the wheel-rail contact point set.
[0084] Embodiment 3
[0085] Based on the above embodiments, Figure 5 This is a functional module diagram of a wheel-rail contact trajectory monitoring system provided by an embodiment of the present disclosure. Figure 5 As shown, a wheel-rail contact track monitoring system comprises:
[0086] The wheel-rail contact track monitoring system 500 described in this embodiment can be installed in an electronic device. According to the functions to be implemented, the wheel-rail contact track monitoring system 500 may include a point cloud measurement module 501, a point cloud segmentation module 502, a point cloud matching module 503, a curve fitting module 504 and a track calculation module 505. The module described in the present disclosure may also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.
[0087] In this embodiment, the functions of each module / unit are as follows:
[0088] The point cloud measurement module 501 is used to perform multi-machine point cloud measurement on the wheel-rail using a preset point cloud sensor to obtain a wheel profile point cloud set, a rail profile point cloud set and a wheel-rail profile point cloud set;
[0089] The point cloud segmentation module 502 is used to split the wheel-rail profile point cloud set into a wheel side point cloud set and a rail side point cloud set;
[0090] The point cloud matching module 503 is used to perform point cloud matching and coordinate transformation on the wheel profile point cloud set using the wheel side point cloud set to obtain a standard wheel point cloud set, and to perform point cloud matching and coordinate transformation on the rail profile point cloud set using the rail side point cloud set to obtain a standard rail point cloud set;
[0091] The curve fitting module 504 is used to fit the standard wheel point cloud set into a wheel profile curve set, and fit the standard rail point cloud set into a rail profile curve set;
[0092] The trajectory calculation module 505 is used to calculate the rail contact point set between the vehicle profile curve set and the rail profile curve set, and generate the wheel-rail contact trajectory according to the rail contact point set.
[0093] In detail, each module described in the wheel-rail contact track monitoring system 500 described in the embodiment of the present disclosure adopts the same technical means as the wheel-rail contact track monitoring method described in Example 1 when used, and can produce the same technical effects, which will not be repeated here.
[0094] Embodiment 4
[0095] On the basis of the above embodiments, this embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the above embodiments.
[0096] In some implementations of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the steps of the method described in the above embodiment are implemented.
[0097] In some implementations of this embodiment, a computer program product is provided, including a computer program / instructions, characterized in that when the computer program is executed by a processor, the steps of the method described in the above embodiment are implemented.
[0098] The processor may include, but is not limited to, one or more processors or microprocessors, etc. Each processor may be an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor or other electronic components to execute the methods in the above embodiments.
[0099] The computer-readable storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, and the computer-readable storage medium may include but is not limited to, for example, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, computer storage media (e.g., hard disk, floppy disk, solid-state drive, removable disk, CD ROM, DVD ROM, Blu-ray disc, etc.).
[0100] The computer-readable storage medium may also store at least one computer executable program, such as a computer-readable instruction. The computer-readable storage medium includes, but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The computer-readable storage medium may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device runs the computer-readable instructions stored on the computer-readable storage medium, the various methods described above may be performed.
[0101] In addition, the computer device may also include (but not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (eg, keyboard, mouse, speaker, etc.), etc.
[0102] The processor may communicate with external devices via an I / O bus via a wired or wireless network.
[0103] In one embodiment, the at least one computer executable instruction may also be compiled into or constitute a software product / computer program product, wherein one or more computer executable instructions are executed by a processor to perform the various functions and / or method steps in the embodiments described in the present technology.
[0104] In the embodiments provided in the present disclosure, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to the multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the above-mentioned module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0105] It should be noted that in the present disclosure, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element limited by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0106] Although the embodiments disclosed in the present disclosure are as above, the above contents are only embodiments adopted for facilitating the understanding of the present disclosure and are not intended to limit the present disclosure. Any technician in the technical field to which the present disclosure belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present disclosure, but the scope of patent protection of the present disclosure shall still be subject to the scope defined in the attached claims.
Claims
1. A method for monitoring wheel-rail contact trajectory, characterized in that: include: Use the preset point cloud sensor to perform multi-machine point cloud measurement on the wheel and rail to obtain the wheel contour point cloud set, rail contour point cloud set and wheel and rail contour point cloud set; Splitting the wheel-rail contour point cloud set into a wheel side point cloud set and a rail side point cloud set; Using the wheel side point cloud set, the wheel contour point cloud set is matched with point clouds and coordinates are transformed to obtain a standard wheel point cloud set; using the rail side point cloud set, the rail contour point cloud set is matched with point clouds and coordinates are transformed to obtain a standard rail point cloud set; Fitting the standard wheel point cloud set into a wheel profile curve set, and fitting the standard rail point cloud set into a rail profile curve set; A wheel-rail contact point set between the vehicle profile curve set and the rail profile curve set is calculated, and a wheel-rail contact trajectory is generated according to the wheel-rail contact point set.
2. The wheel-rail contact trajectory monitoring method according to claim 1, characterized in that: The step of splitting the wheel-rail contour point cloud set into a wheel side point cloud set and a rail side point cloud set comprises: Extracting a wheel-rail point cloud feature set from the wheel-rail contour point cloud set; Performing feature classification on the wheel-rail point cloud feature set to obtain a wheel point cloud feature set and a rail point cloud feature set; Taking the part of the point cloud corresponding to the wheel point cloud feature set in the wheel-rail contour point cloud set as the wheel side point cloud set; The part of point clouds corresponding to the rail point cloud feature set in the wheel-rail contour point cloud set is used as the rail side point cloud set.
3. The wheel-rail contact trajectory monitoring method according to claim 1, characterized in that: The step of performing point cloud matching and coordinate transformation on the wheel contour point cloud set by using the wheel side point cloud set to obtain a standard wheel point cloud set includes: Respectively performing point cloud denoising and point cloud smoothing on the wheel side point cloud set and the wheel contour point cloud set to obtain a denoised wheel side point cloud set and a denoised wheel contour point cloud set; Respectively extracting and encoding feature points from the de-noised wheel side point cloud set and the de-noised wheel contour point cloud set to obtain a wheel side feature point group set and a wheel contour feature point group set; Using the wheel side feature point set to perform point cloud matching on the wheel profile feature point set to obtain a wheel rotation matrix set and a wheel translation matrix set; The wheel rotation matrix set and the wheel translation matrix set are used to perform coordinate transformation on the wheel contour point cloud set to obtain a standard wheel point cloud set.
4. The wheel-rail contact trajectory monitoring method according to claim 3, characterized in that: The step of performing point cloud matching on the wheel profile feature point set using the wheel side feature point set to obtain a wheel rotation matrix set and a wheel translation matrix set includes: Selecting wheel side feature point groups in the wheel side feature point group set one by one as a target side feature point group set, and taking a wheel contour feature point group corresponding to the target side feature point group in the wheel contour feature point group set as a target contour feature point group; Using the target side feature point group to perform feature point registration on the target contour feature point group to obtain a registered wheel side point group; Filtering out the registered wheel contour point group corresponding to the registered wheel side point group from the target contour feature point group; generating a point cloud matching loss model according to the registered wheel side point group and the registered wheel contour point group; Calculating the wheel rotation matrix and the wheel translation matrix corresponding to the registered wheel contour point group according to the point cloud matching loss model; The wheel rotation matrices corresponding to all target side feature point sets in the wheel side feature point set are collected into a wheel rotation matrix set, and the wheel translation matrices corresponding to all target side feature point sets in the wheel side feature point set are collected into a wheel translation matrix set.
5. The wheel-rail contact trajectory monitoring method according to claim 1, characterized in that: The step of fitting the standard wheel point cloud set into a vehicle contour curve set comprises: Selecting standard wheel point clouds in the standard wheel point cloud set one by one as target standard wheel point clouds, and calculating the wheel pair axis of the target standard wheel point cloud; Generating a wheel vertical plane of the target standard wheel point cloud according to the wheel pair axis; Gathering the intersection points of the target standard wheel point cloud and the wheel vertical plane into a wheel fitting point cloud group; The wheel fitting point cloud group is sorted in ascending order of the horizontal coordinate to obtain a wheel fitting point cloud sequence; Performing curve fitting on the wheel fitting point cloud sequence to obtain a wheel contour curve; The vehicle contour curves corresponding to each target standard wheel point cloud in the standard wheel point cloud set are collected into a vehicle contour curve set.
6. The wheel-rail contact trajectory monitoring method according to claim 1, characterized in that: The step of calculating the rail contact point set between the vehicle profile curve set and the rail profile curve set comprises: Selecting vehicle profile curves in the vehicle profile curve set one by one as target vehicle profile curves, and selecting rail profile curves corresponding to the target vehicle profile curves in the rail profile curve set as target rail profile curves; Calculating a vertical distance between the target vehicle profile curve and the target rail profile curve; Performing extreme value calculation on the vertical distance to obtain a vertical distance extreme value group; Screening out a minimum vertical distance extreme value from the vertical distance extreme value group, and performing difference calculation using the minimum vertical distance extreme value and the remaining vertical distance extreme values in the vertical distance extreme value group to obtain a distance difference value group; Determining whether there is a distance difference value in the distance difference value group that is less than a preset wheel-rail contact penetration amount; If not, generating a contact point of the track according to the minimum vertical distance extreme value; If so, the distance difference values of the distance difference group that are smaller than the wheel-rail contact penetration are aggregated into a standard distance difference group, and a contact point with the rail is generated according to the standard distance difference group and the minimum vertical distance extreme value.
7. A wheel-rail contact track monitoring system, characterized in that: include: The point cloud measurement module is used to perform multi-machine point cloud measurement of the wheel and rail using a preset point cloud sensor to obtain a wheel profile point cloud set, a rail profile point cloud set, and a wheel and rail profile point cloud set; A point cloud segmentation module, used for splitting the wheel-rail profile point cloud set into a wheel side point cloud set and a rail side point cloud set; A point cloud matching module, used to perform point cloud matching and coordinate transformation on the wheel profile point cloud set using the wheel side point cloud set to obtain a standard wheel point cloud set, and to perform point cloud matching and coordinate transformation on the rail profile point cloud set using the rail side point cloud set to obtain a standard rail point cloud set; A curve fitting module, used for fitting the standard wheel point cloud set into a wheel profile curve set, and fitting the standard rail point cloud set into a rail profile curve set; The track calculation module is used to calculate the track contact point set between the vehicle profile curve set and the rail profile curve set, and generate the wheel-rail contact track according to the track contact point set.
8. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the wheel-rail contact trajectory monitoring method according to any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for monitoring the wheel-rail contact trajectory described in any one of claims 1 to 6 are implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for monitoring the wheel-rail contact trajectory described in any one of claims 1 to 6 are implemented.