CN-121383895-B - Multi-source data fusion steel rail three-dimensional profile measuring method, device and system

CN121383895BCN 121383895 BCN121383895 BCN 121383895BCN-121383895-B

Abstract

The invention belongs to the technical field of rail traffic detection, and particularly relates to a multi-source data fusion steel rail three-dimensional profile measuring method, device and system, which comprises the following steps of obtaining profile data of a steel rail section, height irregularity data and rail direction irregularity data, and recording mileage information of each measuring position; the method comprises the steps of selecting a local highest point as a vertical splicing reference point, selecting a sampling point abscissa as a transverse splicing reference point, constructing a profile calibration model with smooth constraint, calculating vertical offset and transverse offset of each section by taking high-low irregularity data and track irregularity data as references, and performing global coordinate conversion on original steel rail section profile data at each mileage position according to the calculated vertical offset and transverse offset to generate a continuous steel rail three-dimensional point cloud model with a unified space reference. The invention aims to solve the problem of three-dimensional reconstruction distortion caused by lack of global height and transverse reference in the traditional laser scanning technology.

Inventors

GAO TIANCI
LI LINGXIANG
YAO WEN
WANG YUAN
SHI YIFAN
WANG QILI
LIAO FULIN

Assignees

深圳市埃伯瑞科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251224

Claims (7)

1. A method for measuring a three-dimensional profile of a steel rail by multi-source data fusion is characterized by comprising the following steps: Acquiring profile data of the section of the steel rail at a plurality of discrete mileage positions through a line laser scanning module; Acquiring the height irregularity data and the track direction irregularity data of the steel rail at each mileage position through an irregularity measuring module, and recording the mileage information of each measuring position through a mileage positioning module; selecting a local highest point in the profile of each steel rail section as a vertical splicing reference point; Selecting a sampling point abscissa at a characteristic position of a working edge in a section profile of a steel rail as a transverse splicing reference point; The method comprises the steps of constructing a profile calibration model with smoothness constraint, and calculating the vertical offset and the transverse offset of each section by taking high-low irregularity data and track irregularity data as references, wherein the profile calibration model with smoothness constraint comprises a data fidelity term and a smoothness regularization term, and the reconstructed three-dimensional profile is enabled to not only accord with the irregularity measurement references, but also keep the longitudinal smoothness through optimization solution; Performing global coordinate conversion on the original steel rail section profile data at each mileage position according to the calculated vertical offset and the calculated transverse offset to generate a continuous steel rail three-dimensional point cloud model with a unified space reference; And taking the line laser scanning result as supplementary sensor data, and calculating the height irregularity data and the track irregularity data together with the displacement sensor data in a chord N-point chord measurement method.
2. The method for measuring the three-dimensional profile of the steel rail by fusing multi-source data is characterized in that the global coordinate transformation comprises the steps of conducting coordinate transformation on each profile point in profile data of an original steel rail section according to the vertical offset and the horizontal offset obtained through calculation to obtain a three-dimensional coordinate of the point under a global coordinate system, and splicing and integrating three-dimensional point sets corresponding to all mileage positions according to mileage sequence to form a continuous steel rail three-dimensional point cloud model.
3. A multi-source data fusion rail three-dimensional profile measuring apparatus according to claim 1 or 2, characterized by comprising: The line laser scanning module is used for acquiring profile data of the section of the steel rail at a plurality of discrete mileage positions; The irregularity measuring module is used for acquiring the height irregularity data and the track direction irregularity data of the steel rail at each mileage position through a chord N-point chord measurement method; the mileage positioning module is used for recording mileage information of each measuring position; And the data fusion processing module is connected with the line laser scanning module, the irregularity measuring module and the mileage positioning module and is used for carrying out space-time alignment on the profile data of the section of the steel rail, the height irregularity data and the track direction irregularity data, and calculating the global offset of each section based on the vertical reference point and the transverse reference point so as to realize the reconstruction of the three-dimensional profile of the steel rail.
4. A multi-source data fusion steel rail three-dimensional profile measuring device according to claim 3, wherein the vertical reference point is a local highest point in the steel rail profile, and the transverse reference point is a sampling point abscissa at a characteristic position of a working edge in the steel rail profile.
5. The three-dimensional profile measuring device for the steel rail with multi-source data fusion according to claim 3, wherein the data fusion processing module constructs a profile calibration model with smooth constraint, and the profile calibration model comprises a data fidelity term and a smooth regularization term and is used for optimally solving the vertical offset and the transverse offset of each section.
6. A multi-source data fusion rail three-dimensional profile measuring system comprising a rail three-dimensional profile measuring apparatus as claimed in any one of claims 3 to 5, and: The carrying platform is used for carrying the measuring device to move along the track; The data storage module is connected with the data fusion processing module and used for storing the reconstructed three-dimensional point cloud model of the steel rail; and the display terminal is connected with the data storage module and is used for visually displaying the three-dimensional profile of the steel rail.
7. The multi-source data fusion steel rail three-dimensional profile measuring system according to claim 6, wherein the scanning data of the line laser scanning module and the displacement sensor data of the irregularity measuring module are subjected to data fusion, so that the irregularity data of the track height and the track direction are calculated together, and the irregularity measuring precision is improved.

Description

Multi-source data fusion steel rail three-dimensional profile measuring method, device and system Technical Field The invention belongs to the technical field of rail transit detection, and particularly relates to a method, a device and a system for measuring a three-dimensional profile of a steel rail by multi-source data fusion. Background The three-dimensional profile of the steel rail is one of core parameters for evaluating the geometric state of the rail and guiding maintenance, and the accuracy of the three-dimensional profile directly influences the safety and stability of train operation. In recent years, along with the development of high-speed railways and heavy-load transportation, higher requirements are put on the space continuity, absolute height consistency and detection efficiency of rail profile measurement, and currently, the mainstream rail profile detection method mostly adopts a line laser scanning technology, namely, the rail cross section profile is acquired point by point in the moving process through a line laser sensor arranged on a line detection vehicle, position information is recorded by combining a mileage encoder, and finally a plurality of two-dimensional sections are spliced longitudinally along a rail to form an approximate three-dimensional surface model. The method has the advantages of non-contact, high resolution, high automation degree and the like, and is widely applied to track detection equipment. Problems of the prior art: However, this type of method has a fundamental disadvantage in that the linear laser sensor can only acquire the relative height relationship and the relative lateral relationship between points inside the cross section, and cannot determine the absolute height and the rail position of the cross section in the global vertical coordinate system. Therefore, when a plurality of sections are stacked along the mileage direction, the unified vertical and horizontal references are lacking, so that the problems of height drift, step-like distortion or smoothness distortion and the like of the reconstructed three-dimensional profile exist in the longitudinal direction; conventional rail three-dimensional profile measurement typically relies on a single laser scanner for profile acquisition, but suffers from the following problems: the lack of a longitudinal height and lateral displacement reference between the sections leads to relative height and rail drift; it is difficult to accurately reflect the continuity and smoothness of the track in space; the high-low/rail-direction irregularity information cannot be effectively fused to improve the overall geometric modeling accuracy. Disclosure of Invention The invention aims to provide a method, a device and a system for measuring a three-dimensional profile of a steel rail by multi-source data fusion, and aims to solve the problem of three-dimensional reconstruction distortion caused by lack of global height and transverse reference in the traditional laser scanning technology, and high-precision and high-continuity three-dimensional modeling of the geometric form of the steel rail is realized through an integrated line laser sensor, a high-precision and high-precision irregularity measuring device and a mileage encoder. The technical scheme adopted by the invention is as follows: a multi-source data fusion steel rail three-dimensional profile measuring method comprises the following steps: Acquiring profile data of the section of the steel rail at a plurality of discrete mileage positions through a line laser scanning module; Acquiring the height irregularity data and the track direction irregularity data of the steel rail at each mileage position through an irregularity measuring module, and recording the mileage information of each measuring position through a mileage positioning module; selecting a local highest point in the profile of each steel rail section as a vertical splicing reference point; Selecting a sampling point abscissa at a characteristic position of a working edge in a section profile of a steel rail as a transverse splicing reference point; constructing a profile calibration model with smoothness constraint, and calculating the vertical offset and the transverse offset of each section by taking the height irregularity data and the track irregularity data as references; And carrying out global coordinate conversion on the original steel rail section profile data at each mileage position according to the calculated vertical offset and the calculated transverse offset, and generating a continuous steel rail three-dimensional point cloud model with a unified space reference. According to another aspect of the embodiment of the invention, the profile calibration model with the smoothness constraint comprises a data fidelity term and a smoothness regularization term, and the reconstructed three-dimensional profile is enabled to not only meet the irregularity measurement standard but also kee