CN113158399B - Dynamic splicing processing method and device for rail profile - Google Patents
Dynamic splicing processing method and device for rail profile Download PDFInfo
- Publication number
- CN113158399B CN113158399B CN202011468688.8A CN202011468688A CN113158399B CN 113158399 B CN113158399 B CN 113158399B CN 202011468688 A CN202011468688 A CN 202011468688A CN 113158399 B CN113158399 B CN 113158399B
- Authority
- CN
- China
- Prior art keywords
- rail profile
- steel rail
- profile
- iteration
- splicing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims description 11
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 116
- 239000010959 steel Substances 0.000 claims abstract description 116
- 238000013519 translation Methods 0.000 claims abstract description 65
- 238000000034 method Methods 0.000 claims abstract description 57
- 238000012937 correction Methods 0.000 claims abstract description 37
- 238000004364 calculation method Methods 0.000 claims description 53
- 230000006870 function Effects 0.000 claims description 53
- 238000012545 processing Methods 0.000 claims description 25
- 239000011159 matrix material Substances 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 16
- 230000015572 biosynthetic process Effects 0.000 claims description 14
- 238000003786 synthesis reaction Methods 0.000 claims description 14
- 238000000227 grinding Methods 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 description 28
- 238000010586 diagram Methods 0.000 description 14
- 238000000926 separation method Methods 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000005299 abrasion Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000011282 treatment Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000012804 iterative process Methods 0.000 description 2
- 238000012897 Levenberg–Marquardt algorithm Methods 0.000 description 1
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000011418 maintenance treatment Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000001308 synthesis method Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
Abstract
The invention discloses a method and a device for dynamically splicing steel rail profile, wherein the method comprises the following steps: acquiring an initial value of a steel rail profile splicing parameter; based on the initial value, the rail profile data point corresponding to each cycle is iteratively determined by utilizing an iterative closest point method ICP, and each iteration cycle executes the following operations: according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set; when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period; and when the preset iteration ending condition is met, obtaining the final rail profile according to the rail profile data points corresponding to the iteration period when the preset iteration ending condition is met. The invention can determine the accurate full section profile of the steel rail.
Description
Technical Field
The invention relates to the technical field of steel rail detection, in particular to a method and a device for dynamically splicing steel rail profile.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
The track relationship is a key influencing factor of the safety operation and the comfort of the railway vehicle, and the rail profile is an important component in the track relationship. The detection of the steel rail and the data analysis of the full section profile are all of the importance for guiding safety precaution, maintenance strategies and the like.
At present, the highest detection efficiency of the profile of the railway steel rail is achieved by vehicle-mounted non-contact detection based on a structured light principle, the inner side and the outer side of the steel rail are respectively and dynamically detected by utilizing a laser shooting assembly, and finally, the complete profile is spliced and output as a detection result. Some researches on data fusion, matching and splicing of inner and outer side profiles exist, and laboratory results show that the accuracy is high.
However, in the methods, a set of rotation translation matrix calculation parameters are calculated according to ideal conditions of a laboratory, however, in the actual detection process of the vehicle, the environment is bad, the line conditions are complex, after the laboratory is calibrated, the detection system can generate conditions of inner and outer rail bifurcation, intersection, separation and the like (as shown in fig. 1a to 1 c) of the full-section profile due to various nonlinear errors, especially when the detection system passes through special sections such as a vehicle shaking serious section and a small radius curve, and the real profile cannot be accurately output.
Disclosure of Invention
The embodiment of the invention provides a dynamic splicing processing method for a steel rail profile, which is used for determining an accurate full-section profile of a steel rail, and comprises the following steps:
acquiring an initial value of a steel rail profile splicing parameter;
based on the initial value, determining the steel rail profile data point corresponding to each cycle by utilizing an iterative closest point method ICP iteration, and executing the following operations in each iteration cycle:
according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set;
when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period;
and when the preset iteration ending condition is met, obtaining the final rail profile according to the rail profile data points corresponding to the iteration period when the preset iteration ending condition is met.
The embodiment of the invention also provides a device for dynamically splicing the rail profile, which is used for determining the accurate full-section profile of the rail, and comprises the following components:
the acquisition unit is used for acquiring an initial value of the steel rail profile splicing parameter;
the iteration unit is used for iteratively determining steel rail profile data points corresponding to each cycle by utilizing an iterative closest point method ICP based on the initial value, and each iteration cycle performs the following operations:
according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set;
when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period;
and the determining unit is used for obtaining a final steel rail profile according to steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the rail profile dynamic splicing processing method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium which stores a computer program for executing the rail profile dynamic splicing processing method.
In the embodiment of the invention, in the dynamic splicing processing scheme of the rail profile, compared with the technical scheme that in the prior art, according to ideal conditions of a laboratory, a set of spliced rotation translation matrix calculation parameters are calculated to carry out profile splicing processing, but in the actual detection process of a vehicle, the environment is severe, the line condition is complex, after the laboratory is calibrated, due to various nonlinear errors, a detection system can generate conditions of bifurcation, intersection, separation and the like of inner and outer rails of a full-section profile especially when passing through a serious section of a vehicle and a small radius curve and the like, and the real profile cannot be output correctly, the method is characterized by comprising the following steps: acquiring an initial value of a steel rail profile splicing parameter; based on the initial value, determining the steel rail profile data point corresponding to each cycle by utilizing an iterative closest point method ICP iteration, and executing the following operations in each iteration cycle: according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set; when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period; when the preset iteration ending condition is met, according to the steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met, a final steel rail profile is obtained, and the method is realized:
firstly, the objective function can reduce the weight of a far-end point set when the inner and outer profile matching is carried out, so that the high-precision point set after coordinate conversion is ensured to be used as a main calculation value of the matching, and therefore, certain nonlinear errors on larger nonlinear distortion of the far-end profile obtained by detection are eliminated;
and secondly, performing iterative calculation by utilizing an ICP algorithm, performing intervention on a result after each rotation translation matrix correction, and multiplying a transverse translation parameter by a preset transverse translation correction parameter to serve as a rail profile splicing parameter corresponding to the next iteration period so as to eliminate a transverse translation error caused by the shape characteristic of a rail head region approaching a straight line.
In summary, the rail profile dynamic splicing processing scheme provided by the embodiment of the invention can avoid the conditions of inner and outer rail bifurcation, intersection, separation and the like of the inner profile and the outer profile, and determine the accurate rail full section profile.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
FIGS. 1a to 1c are schematic views of prior art full section profiles showing the bifurcation, crossing, separation, etc. of the inner and outer rails;
FIG. 2 is a schematic flow chart of a method for dynamically splicing rail profiles according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an image with larger interference in an embodiment of the present invention;
fig. 4a to 4b are schematic views showing the missing rail jaw according to the embodiment of the present invention;
FIGS. 5 a-5 b are schematic views showing overexposure or unclear points of the rail jaw according to an embodiment of the present invention;
fig. 6a is an original diagram of the embodiment of the present invention, and fig. 6b is a final splicing result diagram of the embodiment of the present invention;
fig. 7a is an original diagram of the embodiment of the present invention, and fig. 7b is a final splicing result diagram of the embodiment of the present invention;
fig. 8 is a schematic structural diagram of a dynamic splicing processing device for rail profiles in an embodiment of the invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention.
The inventor considers the technical problems existing in the prior art, so a dynamic splicing treatment scheme for the rail profile is provided, and the scheme combines various working conditions possibly encountered in actual operation, so that a rail splicing method is researched, and a rail profile dynamic splicing strategy with more universality and stability is obtained. The following describes the rail profile dynamic splicing treatment scheme in detail.
Fig. 2 is a flow chart of a method for dynamically splicing rail profiles according to an embodiment of the present invention, as shown in fig. 2, the method includes the following steps:
step 101: acquiring an initial value of a steel rail profile splicing parameter;
step 102: based on the initial value, determining the steel rail profile data point corresponding to each cycle by utilizing an iterative closest point method ICP iteration, and executing the following operations in each iteration cycle:
step 1021: according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set;
step 1022: when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period;
step 103: and when the preset iteration ending condition is met, obtaining the final rail profile according to the rail profile data points corresponding to the iteration period when the preset iteration ending condition is met.
In the embodiment of the invention, in the dynamic splicing processing scheme of the rail profile, compared with the technical scheme that in the prior art, according to ideal conditions of a laboratory, a set of spliced rotation translation matrix calculation parameters are calculated to carry out profile splicing processing, but in the actual detection process of a vehicle, the environment is severe, the line condition is complex, after the laboratory is calibrated, due to various nonlinear errors, the detection system can generate conditions of bifurcation, intersection, separation and the like of inner and outer rails of a full-section profile especially when passing through a serious section of a vehicle, a small radius curve and other special sections, and the real profile cannot be output correctly, the method has the advantages that:
firstly, the objective function can reduce the weight of a far-end point set when the inner and outer profile matching is carried out, so that the high-precision point set after coordinate conversion is ensured to be used as a main calculation value of the matching, and therefore, certain nonlinear errors on larger nonlinear distortion of the far-end profile obtained by detection are eliminated;
and secondly, performing iterative calculation by utilizing an ICP algorithm, performing intervention on a result after each rotation translation matrix correction, and multiplying a transverse translation parameter by a preset transverse translation correction parameter to serve as a rail profile splicing parameter corresponding to the next iteration period so as to eliminate a transverse translation error caused by the shape characteristic of a rail head region approaching a straight line.
In summary, the dynamic splicing processing scheme of the rail profile provided by the embodiment of the invention can avoid the conditions of inner and outer rail bifurcation, intersection, separation and the like of the inner and outer profiles, and determine the accurate full-section rail profile.
In practice, the condition for ending the preset iteration may be that the value of the ICP objective function is less than a small value, such as 0.005.
In one embodiment, in the step 101, obtaining the initial value of the rail profile splicing parameter may include: and calibrating the left and right profile synthesis of the same calibration block through a laboratory, obtaining a calibration rotation translation matrix parameter of the same steel rail inner and outer profile synthesis, and taking the calibration rotation translation matrix parameter as an initial value of the steel rail profile splicing parameter.
In specific implementation, the method for acquiring the initial value of the steel rail profile splicing parameter is beneficial to further improving the accuracy of the steel rail profile.
In one embodiment, after the step 101, the method for dynamically splicing the rail profile may further include: selecting a preset area from the steel rail profile as a calculation area;
thus, in the subsequent step 102, iteratively determining rail profile data points corresponding to each cycle using iterative closest point method ICP based on the initial values may include: and based on the initial value, iteratively determining the steel rail profile data point corresponding to each period in the calculation domain by utilizing an iterative closest point method ICP.
In specific implementation, a calculation domain is determined, and based on the initial value, the steel rail profile data points corresponding to each period in the calculation domain are determined through iteration by utilizing an iterative closest point method ICP, so that the calculated amount is reduced, part of interference points are removed, and the efficiency and the accuracy of the steel rail profile dynamic splicing processing are further improved.
In one embodiment, selecting a preset region from the rail profile as the calculation domain may include:
determining coordinate position information of the top points and the side grinding points of the inner profile rail and the outer profile rail;
and determining the calculation domain as a rectangle, and determining the position information of the rectangular calculation domain according to the coordinate position information of the vertices of the inner profile rail and the side grinding points.
In specific implementation, the above-mentioned implementation of selecting the calculation domain further improves the efficiency and precision of the dynamic splicing processing of the rail profile.
In one embodiment, the outer side rail vertex and side grinding point coordinates are respectivelyThe coordinates of the vertex and the side grinding point of the inner side rail are respectively +.>
The location information of the rectangular computation domain includes: the upper left endpoint coordinates of the calculated domain diagonal are: the lower right endpoint coordinates of the diagonal of the calculation domain are: />Wherein: delta 1 、δ 2 、δ 3 And delta 4 The range of the values of (2) to (5) is all 2.
In specific implementation, the specific calculation domain information further improves the efficiency and the precision of the dynamic splicing processing of the steel rail profile.
In one embodiment, in the step 1021, the objective function may be:
wherein the intermediate part point set is selected asAll points in the computation domain in the interval.
In specific implementation, the objective function can eliminate certain nonlinear errors which still exist in the correction of the distortion by the calibration method model, so that the objective function reduces the weight of a far-end point set during inner and outer profile matching, and further improves the accuracy of the steel rail profile.
In one embodiment, in the step 1022, the method of dynamic splicing of rail profiles may further include determining the lateral translation correction parameter according to the following method:
determining the matched jaw track distance according to the track profile images with the preset number of definition larger than a preset value;
determining a probability density function according to the jaw distance;
constructing a likelihood function according to the probability density function;
and determining the optimal transverse translation correction parameters according to the likelihood function.
In specific implementation, the embodiment of determining the transverse translation correction parameters is helpful for further improving the more accurate and real steel rail profile.
In one embodiment, in step 103, when the preset iteration end condition is satisfied, obtaining the final rail profile according to the rail profile data points corresponding to the iteration cycle when the preset iteration end condition is satisfied may include:
when the preset iteration ending condition is met, steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met are obtained;
screening steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met, and obtaining screened steel rail profile data points;
and obtaining the final steel rail profile according to the screened steel rail profile data points.
In specific implementation, after screening treatment is carried out on steel rail profile data points corresponding to iteration periods when preset iteration ending conditions are met, a final steel rail profile is obtained according to the screened steel rail profile data points, and the accuracy of the steel rail profile is further improved.
The following description is given by way of example to facilitate understanding of how the invention may be practiced.
The rail profile dynamic splicing processing method provided by the embodiment of the invention comprises the following steps:
(1) Confirming initial value
The calibration of the left and right profile synthesis is carried out on the same calibration block through a laboratory, a calibration rotation translation matrix [ R, t ] of the same steel rail inner and outer profile synthesis is obtained, and the parameter is used as an initial value of dynamic profile splicing;
the splicing parameters obtained through laboratory calibration can realize millimeter-level registration of rail profile splicing, but the high-precision requirement (within 0.2 mm) of a system is difficult to reach, and the nonlinear errors caused by factors such as different imaging positions do not have good adaptability, so that the laboratory calibration synthesis parameters are required to be used as initial values for further processing of subsequent profile splicing.
(2) Selecting a computational domain
Without loss of generality, the following method description takes the left rail as an example:
the inventor considers one technical problem: in order to improve the calculation efficiency and eliminate the influence of non-railhead interference (such as dirt, vegetation, etc.) in the image as much as possible, a calculation domain as small as possible (the size of a preset area) needs to be selected.
Calculating the vertex and the side grinding point of the inner profile rail: according to the definition, the rail vertex is the highest position (the positions indicated by the two small o at the top in fig. 3) of the profile rail top surface, the side grinding point (also called the gauge point) is the point 16mm below the rail top (the positions indicated by the two small o at the side in fig. 3), and according to the definition, the corresponding point can be obtained by carrying out statistical calculation on the point set after the extraction of the image light bar. The outer rail top and the side grinding point are marked asThe inside rail top and the side grinding point seat are marked as +.>
In the embodiment of the invention, the calculation domain is selected as a rectangle (mainly calculating the rail top position of the upper half of the rail head), and the outside of the rail is considered to be not contacted with the wheel, so that the abrasion is negligible, and the left upper endpoint coordinate of the diagonal line of the calculation domain is:the coordinates of the right lower endpoint of the diagonal of the calculation domain are: /> Wherein delta 1 ～δ 4 The smaller number, typically 2-5 pixels, is chosen to avoid the problem of the relevant boundaries caused by computing the domain boundaries just past the valid computation points.
The selection of the calculation domain has positive effects on the matching of images with serious partial dirt and stray light interference, as shown in fig. 2, a camera on one side has a plurality of spots due to dirt, and the calculation amount is reduced and partial interference points are removed through the selection of the calculation domain.
(3) Inner and outer matching objective function
In the world coordinate system, the sets of inner and outer profile points within the computational domain are denoted as P (P 1 ,p 2 ,p 3 …) and Q (Q) 1 ,q 2 ,q 3 …). These two sets of points are actually detected image data for the same location of the rail, so that matching can be performed using the iterative closest point method (ICP). The inventor considers one technical problem: in the actual detection results, the inner profile and the outer profile often have bifurcation, intersection and other conditions, one of the main reasons is that the far-end profile detected by each side camera has larger nonlinear distortion (compared with the near end), and the calibration method model used at present still has a certain correction for the distortionNonlinear errors, therefore, the objective function needs to reduce the weight of the far end point (the far end point refers to the point far from the rail vertex and gauge point in the measured half-section profile) set when performing the inner and outer profile matching (the weight is represented by the second part in the following formula as N in the "intermediate point set" C The calculation of the point pairs is equivalent to the calculation of the deviation of the point sets again so as to ensure that the high-precision point sets after coordinate conversion are used as main calculation values of matching. The middle part point set of the inner and outer profile is marked as A (a 1 ,a 2 ,a 3 …) and B (B) 1 ,b 2 ,b 3 …), the objective function of ICP algorithm matching is:
wherein the intermediate part point set (the point set of the coincident part near the rail top point in the two half-section profiles, namely the point set of the coincident part on the abscissa) is selected as the specific selection method as described in the following (4) iterative process optimizationPoints in all calculation fields in the interval are points in the inner and outer profiles, p and q are points in the inner and outer profiles, a and b are points in the middle part point set of the inner and outer profiles, R and T are rotation and translation matrices, N represents the total number of point sets, and NC represents the total number of middle part point sets.
(4) Iterative process optimization
The invention also finds a technical problem: the actual rail profile has a large radius of the curve section near the rail head area, and the geometric characteristics of the rail head area tend to be straight under the influence of external forces such as long-term rail abrasion and the like. In addition, in the section with serious rail abrasion, in order to prevent further development of surface cracks, the working department can mill and grind the surface of the rail, and after similar working maintenance treatment, the rail head part of the rail is directly in a straight line. For this practical situation, since the weights of the intermediate partial point sets are larger in the objective function, the convergence result of the horizontal translation parameters in the translation matrix in the actual iterative matching process may be more dependent on the horizontal distribution of the points after the extraction of the image light bar, but less related to the actual shape of the rail, so that the rail is "narrowed" or "widened", which causes a larger interference to the subsequent analysis of the rail profile.
The inventor comprehensively considers that one of main reasons for the conditions that the inner and outer detection profiles of the steel rail are bifurcated is an error when the actual image of the steel rail is extracted into the profile with one pixel width, and the error can cause the inner and outer profiles of the steel rail to be separated in the vertical direction, so that the horizontal position is hardly interfered. Therefore, the embodiment of the invention considers the objective function in the step (3), intervenes in the result after each rotation translation matrix correction when the ICP algorithm is utilized to carry out iterative computation, and multiplies the transverse translation parameter by a coefficient xi (transverse translation correction parameter) smaller than 1 as the initial value of the next computation to eliminate the transverse translation error caused by the shape characteristic of the rail head area approaching to the straight line.
(5) Determination of lateral translation correction parameter ζ
The invention also finds a technical problem: the calculation errors of the transverse translation parameters are derived from the distribution of light bar extraction points, the shape characteristics of a rail head area, the unsmooth distribution of a line, the random change of the position and the posture caused by vehicle vibration and other systematic errors, and the calculation errors of the transverse translation parameters caused by the error sources are all independently and uniformly distributed, so that the embodiment of the invention utilizes maximum likelihood estimation to select a Gaussian distribution model for calculation when the parameters are selected.
The invention also finds that: the inner and outer rail jaws (the rail jaws refer to the positions of the centers of small circles of the left and right lower corners R2.5 of the rail head of the rail, and R2.5 means that the arc section with the radius of 2.5 of the left and right lower corners in a rail profile standard diagram (each rail profile has national standards) are not directly involved in wheel-rail contact, so that abrasion is hardly caused, and the distance between the rail jaws of the inner and outer rails can be used as a reference basis for judging the transverse translation of the splicing of the rails. The influence of dirt, railway ballasts, vehicle position and posture and the like in an actual line can cause partial deletion of a rail jaw image during rail profile detection (see fig. 4a to 4 b), and meanwhile, the shape characteristics of the rail jaw part can cause overexposure or unclear rail jaw part of a partial profile image (see fig. 5a to 5 b), so that the rail jaw part cannot be used as a transverse reference basis during dynamic matching, but a complete profile containing clear rail jaws can be selected to calculate transverse translation correction parameters in the step. The specific determination flow comprises the following steps:
selecting N d (200-500) clear rail profile images, running the matching algorithm, and calculating the matched rear rail jaw distance D, wherein the rail jaw distance is a determined value D for the same rail profile.
The probability density function is:
constructing a likelihood function:
let L take the maximum value, let the following minimum:
wherein d represents the actual calculated jaw distance; d represents the jaw distance of the theoretical profile; sigma is the standard deviation and ζ is the transverse matching correction parameter.
The most suitable correction value can be obtained by using a Levenberg-Marquardt algorithm for solving the optimization problem of the multi-parameter nonlinear system.
Consider detecting an operation mode of a train: each road bureau prepares the detection vehicle to detect the corresponding route according to the established plan, and the detection mode has repeatability, so that the correction parameters of the same vehicle and the same route are calculated, and further, a simple correction parameter library can be automatically built for different vehicles and routes, so that the correction method has great significance.
(6) Final rail profile
After the steps (1) - (6), the profile splicing is completed, and the invention also considers that: proper screening of the inside and outside data points is also required as a final output profile. For the reasons described in step (3), the non-middle parts of the inner and outer sides of the final profile are directly selected from the detection results of the cameras on the corresponding sides. In the case of similar engineering, mathematical models such as fitting and the like are mostly adopted for fusion calculation, but the system is considered as a detection system, if fitting is carried out on middle true points, human errors caused by fitting are uncontrollable, so that the embodiment of the invention fully reserves the middle point set when determining the final profile, and the points have important significance for ensuring the precision of the detection system. And finally, taking the data point set as a final profile to output, and carrying out corresponding further processing according to the detected point set when carrying out profile analysis and corresponding calculation.
The final splice result is shown in fig. 6a to 7b, wherein fig. 6a and 7a are original drawings, and fig. 6b and 7b are post-splice results (final rail profile).
In order to achieve dynamic synthesis of the inner and outer profiles in driving situations, a method with strong adaptability needs to be proposed, so that many reasons possibly causing synthesis failure must be avoided, for example, the aforementioned lateral position of the synthesized profile is controlled by using the width of the rail jaw. The embodiment of the invention can realize high-precision synthesis of the profile after various interferences.
In summary, the rail profile dynamic splicing processing scheme provided by the embodiment of the invention realizes:
(1) In order to realize dynamic synthesis of the rail profile, a synthesis method with universality is adopted by combining engineering practice, synthesis calculation can be completed under various interferences and complex environments, and higher precision can be realized;
(2) The smaller calculation domain is determined by utilizing the track gauge points at the track top, so that the algorithm searching and calculating time is greatly improved;
(3) In consideration of engineering practice, different parts of the profile of the steel rail are given different weights (secondary calculation is carried out on the middle point set in the objective function and the later calculation of the transverse offset parameter), and the middle point set with higher accuracy is taken as an important basis for matching;
(4) The geometric characteristics of the inner and outer splicing of the steel rail and the mode method in the actual operation of the working department are fully considered, the transverse translation matrix in the synthetic matrix is corrected, and a method is provided for determining correction parameters;
(5) Based on the detection purpose and the full understanding of the profile detection system, a simple scheme is provided for the choice and the rejection of the synthesized profile point set, human errors caused by traditional mathematical fitting and the like are abandoned, and a real measured value is reserved as a detection result.
The embodiment of the invention also provides a rail profile dynamic splicing processing device, which is described in the following embodiment. Because the principle of the device for solving the problems is similar to that of the rail profile dynamic splicing processing method, the implementation of the device can be referred to the implementation of the rail profile dynamic splicing processing method, and the repetition is not repeated.
Fig. 8 is a schematic structural diagram of a dynamic splicing processing device for rail profiles according to an embodiment of the present invention, as shown in fig. 8, the device includes:
an obtaining unit 02, configured to obtain an initial value of a rail profile splicing parameter;
the iteration unit 04 is configured to iteratively determine, based on the initial value, a steel rail profile data point corresponding to each cycle by using an iterative closest point method ICP, where each iteration cycle performs the following operations:
according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set;
when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period;
and the determining unit 06 is used for obtaining a final rail profile according to the rail profile data points corresponding to the iteration period when the preset iteration ending condition is met.
In an embodiment, the determining unit is specifically configured to:
when the preset iteration ending condition is met, steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met are obtained;
screening steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met, and obtaining screened steel rail profile data points;
and obtaining the final steel rail profile according to the screened steel rail profile data points.
In one embodiment, the objective function is:
wherein the intermediate part point set is selected asAll points in the computation domain in the interval.
In one embodiment, the obtaining unit is specifically configured to: and calibrating the left and right profile synthesis of the same calibration block through a laboratory, obtaining a calibration rotation translation matrix parameter of the same steel rail inner and outer profile synthesis, and taking the calibration rotation translation matrix parameter as an initial value of the steel rail profile splicing parameter.
In one embodiment, the above-mentioned rail profile dynamic splicing processing device may further include: the calculation domain determining unit is used for selecting a preset area from the steel rail profile as a calculation domain;
the iteration unit is specifically configured to: and based on the initial value, iteratively determining the steel rail profile data point corresponding to each period in the calculation domain by utilizing an iterative closest point method ICP.
In one embodiment, the above-mentioned calculation domain determining unit is specifically configured to:
determining coordinate position information of the top points and the side grinding points of the inner profile rail and the outer profile rail;
and determining the calculation domain as a rectangle, and determining the position information of the rectangular calculation domain according to the coordinate position information of the vertices of the inner profile rail and the side grinding points.
In one embodiment, the above-mentioned rail profile dynamic splicing processing device may further include: a lateral translation correction parameter determining unit configured to determine the lateral translation correction parameter according to the following method:
determining the matched jaw track distance according to the track profile images with the preset number of definition larger than a preset value;
determining a probability density function according to the jaw distance;
constructing a likelihood function according to the probability density function;
and determining the optimal transverse translation correction parameters according to the likelihood function.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the rail profile dynamic splicing processing method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium which stores a computer program for executing the rail profile dynamic splicing processing method.
In the embodiment of the invention, in the dynamic splicing processing scheme of the rail profile, compared with the technical scheme that in the prior art, according to ideal conditions of a laboratory, a set of spliced rotation translation matrix calculation parameters are calculated to carry out profile splicing processing, but in the actual detection process of a vehicle, the environment is severe, the line condition is complex, after the laboratory is calibrated, due to various nonlinear errors, a detection system can generate conditions of bifurcation, intersection, separation and the like of inner and outer rails of a full-section profile especially when passing through a serious section of a vehicle and a small radius curve and the like, and the real profile cannot be output correctly, the method is characterized by comprising the following steps: acquiring an initial value of a steel rail profile splicing parameter; based on the initial value, determining the steel rail profile data point corresponding to each cycle by utilizing an iterative closest point method ICP iteration, and executing the following operations in each iteration cycle: according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set; when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period; when the preset iteration ending condition is met, according to the steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met, a final steel rail profile is obtained, and the method is realized:
firstly, the objective function can reduce the weight of a far-end point set when the inner and outer profile matching is carried out, so that the high-precision point set after coordinate conversion is ensured to be used as a main calculation value of the matching, and therefore, certain nonlinear errors on larger nonlinear distortion of the far-end profile obtained by detection are eliminated;
and secondly, performing iterative calculation by utilizing an ICP algorithm, performing intervention on a result after each rotation translation matrix correction, and multiplying a transverse translation parameter by a preset transverse translation correction parameter to serve as a rail profile splicing parameter corresponding to the next iteration period so as to eliminate a transverse translation error caused by the shape characteristic of a rail head region approaching a straight line.
In summary, the rail profile dynamic splicing processing scheme provided by the embodiment of the invention can avoid the conditions of inner and outer rail bifurcation, intersection, separation and the like of the inner profile and the outer profile, and determines the accurate rail profile.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (9)
1. A rail profile dynamic splicing processing method is characterized by comprising the following steps:
acquiring an initial value of a steel rail profile splicing parameter;
based on the initial value, determining the steel rail profile data point corresponding to each cycle by utilizing an iterative closest point method ICP iteration, and executing the following operations in each iteration cycle:
according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set;
when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period; wherein the lateral translation correction parameter is determined as follows: determining the matched jaw track distance according to the track profile images with the preset number of definition larger than a preset value; determining a probability density function according to the jaw distance; constructing a likelihood function according to the probability density function; determining an optimal transverse translation correction parameter according to the likelihood function;
and when the preset iteration ending condition is met, obtaining the final rail profile according to the rail profile data points corresponding to the iteration period when the preset iteration ending condition is met.
2. The method for dynamically splicing the rail profile according to claim 1, wherein when a preset iteration end condition is satisfied, obtaining a final rail profile according to rail profile data points corresponding to an iteration cycle when the preset iteration end condition is satisfied, comprises:
when the preset iteration ending condition is met, steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met are obtained;
screening steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met, and obtaining screened steel rail profile data points;
and obtaining the final steel rail profile according to the screened steel rail profile data points.
3. The method for dynamically splicing rail profiles according to claim 1, wherein the objective function is:
wherein p and q are points in the inner and outer profiles, a and b are points in the middle part of the inner and outer profiles, R and T are rotation matrix and translation matrix, N represents total point set, N C Representing the total number of mid-portion point sets.
4. The method for dynamically splicing the rail profile according to claim 1, wherein obtaining the initial value of the splicing parameter of the rail profile comprises: and calibrating the left and right profile synthesis of the same calibration block through a laboratory, obtaining a calibration rotation translation matrix parameter of the same steel rail inner and outer profile synthesis, and taking the calibration rotation translation matrix parameter as an initial value of the steel rail profile splicing parameter.
5. The method for dynamically splicing rail profiles according to claim 1, further comprising: selecting a preset area from the steel rail profile as a calculation area;
based on the initial value, determining the steel rail profile data point corresponding to each period by utilizing an iterative closest point method ICP in an iterative manner, wherein the method comprises the following steps: and based on the initial value, iteratively determining the steel rail profile data point corresponding to each period in the calculation domain by utilizing an iterative closest point method ICP.
6. The method for dynamically splicing steel rail profiles according to claim 5, wherein selecting a preset area from the steel rail profiles as a calculation area comprises:
determining coordinate position information of the top points and the side grinding points of the inner profile rail and the outer profile rail;
and determining the calculation domain as a rectangle, and determining the position information of the rectangular calculation domain according to the coordinate position information of the vertices of the inner profile rail and the side grinding points.
7. A rail profile dynamic splicing processing device, comprising:
the acquisition unit is used for acquiring an initial value of the steel rail profile splicing parameter;
the iteration unit is used for iteratively determining steel rail profile data points corresponding to each cycle by utilizing an iterative closest point method ICP based on the initial value, and each iteration cycle performs the following operations:
according to the steel rail profile splicing parameters of the current period and a pre-established ICP matching objective function, obtaining steel rail profile data points corresponding to the current period; the objective function is used for reducing the weight of the far-end profile point set;
when the preset iteration ending condition is not met, multiplying the transverse translation parameter corresponding to the steel rail profile of the current period by the preset transverse translation correction parameter to serve as the steel rail profile splicing parameter corresponding to the next iteration period; wherein the lateral translation correction parameter is determined as follows: determining the matched jaw track distance according to the track profile images with the preset number of definition larger than a preset value; determining a probability density function according to the jaw distance; constructing a likelihood function according to the probability density function; determining an optimal transverse translation correction parameter according to the likelihood function;
and the determining unit is used for obtaining a final steel rail profile according to steel rail profile data points corresponding to the iteration period when the preset iteration ending condition is met.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 6 when executing the computer program.
9. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program for executing the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011468688.8A CN113158399B (en) | 2020-12-14 | 2020-12-14 | Dynamic splicing processing method and device for rail profile |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011468688.8A CN113158399B (en) | 2020-12-14 | 2020-12-14 | Dynamic splicing processing method and device for rail profile |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113158399A CN113158399A (en) | 2021-07-23 |
| CN113158399B true CN113158399B (en) | 2024-03-12 |
Family
ID=76882555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011468688.8A Active CN113158399B (en) | 2020-12-14 | 2020-12-14 | Dynamic splicing processing method and device for rail profile |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113158399B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113806867B (en) * | 2021-09-16 | 2024-02-02 | 中国国家铁路集团有限公司 | Equivalent taper correction method and device based on actual measurement profile matching |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105115976A (en) * | 2015-06-24 | 2015-12-02 | 上海图甲信息科技有限公司 | Detection system and method for wear defects of track |
| CN105783779A (en) * | 2016-04-28 | 2016-07-20 | 湖南大学 | Steel rail contour real-time form identification and distortion calibration method based on three-layer matching |
| CN108955576A (en) * | 2018-10-31 | 2018-12-07 | 湖南东映碳材料科技有限公司 | Multi-line structured light self-calibrating method and system in profile of steel rail dynamic detection |
| CN109798844A (en) * | 2018-12-28 | 2019-05-24 | 北京交通大学 | A kind of rail exterior feature autoegistration method in rail wear dynamic measurement |
| CN110688710A (en) * | 2019-09-27 | 2020-01-14 | 南京理工大学 | Turning repairing method based on rail transit vehicle wheel pair service life statistical model |
| CN110866309A (en) * | 2019-11-15 | 2020-03-06 | 中车南京浦镇车辆有限公司 | Subway clearance-based intelligent calculation method for maximum contour of metro vehicle |
| CN111625920A (en) * | 2019-09-03 | 2020-09-04 | 东莞灵虎智能科技有限公司 | Steel rail profile intelligent analysis method based on hidden Markov chain model |
| CN111710027A (en) * | 2020-05-25 | 2020-09-25 | 南京林业大学 | Tunnel three-dimensional geometric reconstruction method considering data-driven segment segmentation and model-driven segment assembly |
| CN112037211A (en) * | 2020-09-04 | 2020-12-04 | 中国空气动力研究与发展中心超高速空气动力研究所 | Damage characteristic identification method for dynamically monitoring small space debris impact event |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI403690B (en) * | 2009-10-26 | 2013-08-01 | Ind Tech Res Inst | Self-localizing device and method thereof |
-
2020
- 2020-12-14 CN CN202011468688.8A patent/CN113158399B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105115976A (en) * | 2015-06-24 | 2015-12-02 | 上海图甲信息科技有限公司 | Detection system and method for wear defects of track |
| CN105783779A (en) * | 2016-04-28 | 2016-07-20 | 湖南大学 | Steel rail contour real-time form identification and distortion calibration method based on three-layer matching |
| CN108955576A (en) * | 2018-10-31 | 2018-12-07 | 湖南东映碳材料科技有限公司 | Multi-line structured light self-calibrating method and system in profile of steel rail dynamic detection |
| CN109798844A (en) * | 2018-12-28 | 2019-05-24 | 北京交通大学 | A kind of rail exterior feature autoegistration method in rail wear dynamic measurement |
| CN111625920A (en) * | 2019-09-03 | 2020-09-04 | 东莞灵虎智能科技有限公司 | Steel rail profile intelligent analysis method based on hidden Markov chain model |
| CN110688710A (en) * | 2019-09-27 | 2020-01-14 | 南京理工大学 | Turning repairing method based on rail transit vehicle wheel pair service life statistical model |
| CN110866309A (en) * | 2019-11-15 | 2020-03-06 | 中车南京浦镇车辆有限公司 | Subway clearance-based intelligent calculation method for maximum contour of metro vehicle |
| CN111710027A (en) * | 2020-05-25 | 2020-09-25 | 南京林业大学 | Tunnel three-dimensional geometric reconstruction method considering data-driven segment segmentation and model-driven segment assembly |
| CN112037211A (en) * | 2020-09-04 | 2020-12-04 | 中国空气动力研究与发展中心超高速空气动力研究所 | Damage characteristic identification method for dynamically monitoring small space debris impact event |
Non-Patent Citations (2)
| Title |
|---|
| 基于双重匹配参数估计的钢轨动态轮廓校准方法;李艳福 等;《铁道学报》(第03期);112-119 * |
| 钢轨轮廓全断面检测中的匹配方法研究;朱新荣 等;《铁道科学与工程学报》;第15卷(第09期);2401-2406 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113158399A (en) | 2021-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2743020C2 (en) | System and method for monitoring railway wheels | |
| CN111274671B (en) | Precise repair method for complex product assembly process based on digital twin and operation system thereof | |
| CN102159918B (en) | Method and measuring assembly for determining wheel or axle geometry of vehicle | |
| JP7265592B2 (en) | Techniques for Measuring Overlay Between Layers in Multilayer Structures | |
| CN107607060A (en) | A kind of phase error compensation method in the measurement applied to grating tripleplane | |
| CN102589435B (en) | Efficient and accurate detection method of laser beam center under noise environment | |
| US20130258060A1 (en) | Information processing apparatus that performs three-dimensional shape measurement, information processing method, and storage medium | |
| CN108053441B (en) | High-precision measurement method by using laser triangulation method | |
| CN104024792A (en) | Tire shape inspection method and tire shape inspection device | |
| CN103678754A (en) | Information processing apparatus and information processing method | |
| CN115482195B (en) | Train part deformation detection method based on three-dimensional point cloud | |
| CN113158399B (en) | Dynamic splicing processing method and device for rail profile | |
| CN114577131B (en) | Vehicle body gap detection method and system based on 3D structured light camera | |
| CN112815843B (en) | On-line monitoring method for printing deviation of workpiece surface in 3D printing process | |
| CN113643372B (en) | Three-dimensional weld extraction method and system | |
| CN112161586A (en) | Line structured light vision sensor calibration method based on coding checkerboard | |
| CN115546125A (en) | Method for error detection and track deviation correction of additive manufacturing cladding layer based on point cloud information | |
| CN114612412A (en) | Processing method of three-dimensional point cloud data, application of processing method, electronic device and storage medium | |
| CN116128907A (en) | Open wagon side wall overhauling accurate positioning cutting method based on laser scanning image | |
| JP6951469B2 (en) | How to calibrate an optical measuring device | |
| CN115205397A (en) | Vehicle space-time information identification method based on computer vision and pose estimation | |
| CN109902694B (en) | Extraction method of square hole characteristics | |
| CN111462216B (en) | Method for determining circle center pixel coordinates in circular array calibration plate | |
| CN113776456A (en) | Curve section steel rail outline measurement error correction method and device based on double-line laser | |
| CN113312715A (en) | Tramcar groove rail distortion irregularity prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |