Vehicle body calibration method, device, equipment and program product of rail traffic tool

Abstract

The invention relates to the field of rail transit, in particular to a rail transit vehicle body calibration method, device and equipment and a program product. The method comprises the following steps: acquiring point cloud data of a track environment in front of a train through a point cloud acquisition device, and extracting a track surface point cloud and a track inner edge point cloud according to the point cloud data; fitting according to the track surface point cloud to obtain a track surface normal vector, and fitting according to the track inner edge point cloud to obtain a track line vector; determining a rotation matrix from a point cloud acquisition device coordinate system to a vehicle body coordinate system according to the orbital plane normal vector and the orbital line vector; determining a translation vector from a point cloud acquisition device coordinate system to a vehicle body coordinate system according to the rotation matrix and the track inner edge point cloud; and completing calibration of the vehicle body coordinate system according to the rotation matrix and the translation vector. According to the method, a special calibration site does not need to be built, the calibration cost is low, a specific calibration object does not need to be placed, and the calibration efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of rail transit, and in particular to a rail vehicle body calibration method, device, equipment and program product. BACKGROUND

[0002] Laser radar has become a key sensor of a train automatic driving perception system due to its high-precision three-dimensional measurement capability. Integrating sensors such as laser radar at the front end of the train can collect real-time front environment data and realize autonomous perception function. The accuracy of the train perception system highly depends on the calibration accuracy of the vehicle body, that is, accurately determining the relative position and attitude relationship between the laser radar and the vehicle body.

[0003] Currently, when calibrating the vehicle body, there are special requirements for the calibration environment, such as requiring a wall perpendicular to the ground in front of the vehicle body, or requiring a specially-made calibration object, such as placing a specific calibration board. Due to the large volume of the train and the special running environment, if a special calibration site is built for vehicle body calibration, the calibration cost is high, and when a specific calibration object is placed for vehicle body calibration, the operation efficiency of the vehicle body calibration is low. SUMMARY

[0004] Therefore, the embodiments of the present application provide a rail vehicle body calibration method, device, equipment and program product to solve the problems of high calibration cost or low calibration operation efficiency in the prior art.

[0005] A first aspect of the embodiments of the present application provides a rail vehicle body calibration method, and the method comprises:

[0006] acquiring point cloud data of a track environment in front of a train, extracting track surface point cloud and track inside point cloud according to the point cloud data;

[0007] fitting according to the track surface point cloud to obtain a track surface normal vector, and fitting according to the track inside point cloud to obtain a track line vector;

[0008] determining a rotation matrix from a point cloud acquisition device coordinate system to a vehicle body coordinate system according to the track surface normal vector and the track line vector;

[0009] determining a translation vector from the point cloud acquisition device coordinate system to the vehicle body coordinate system according to the rotation matrix and the track inside point cloud;

[0010] completing calibration of the vehicle body coordinate system according to the rotation matrix and the translation vector.

[0011] In combination with the first aspect, in a first possible implementation manner of the first aspect, the extracting the track surface point cloud and the track inside point cloud according to the point cloud data comprises:

[0012] Determine a region of interest according to the track width and the train height;

[0013] Perform point cloud segmentation on the point cloud data in the region of interest to obtain left track point cloud and right track point cloud;

[0014] Project the left track point cloud and the right track point cloud onto a first plane to obtain a first plane projected point cloud distribution, the first plane being an XOZ plane of a point cloud collection device coordinate system, the center O of the point cloud collection device coordinate system being a center position of the point cloud collection device, the point cloud collection device coordinate system complying with the right-hand rule, and simultaneously defining the X axis as a direction from the center of the point cloud collection device coordinate system outward and perpendicular to the emission plane, with the right-hand index finger direction as the X axis direction, the middle finger direction as the Y axis, and the thumb direction as the Z axis;

[0015] Extract the track face point cloud according to the first plane projected point cloud distribution, and extract the track inside along point cloud based on the track face point cloud.

[0016] In a second possible implementation manner of the first aspect, in combination with the first possible implementation manner of the first aspect, the track face point cloud comprises left track face point cloud and right track face point cloud;

[0017] Extracting the track face point cloud according to the first plane projected point cloud distribution comprises:

[0018] Analyzing the first plane projected point cloud distribution from top to bottom along the reverse direction of the Z axis to determine the first plane projected point cloud distribution density, and determining a first projection region with the first plane projected point cloud distribution density greater than a predetermined density threshold;

[0019] Extracting the left track face point cloud and the right track face point cloud projected to the first projection region.

[0020] In a third possible implementation manner of the first aspect, in combination with the second possible implementation manner of the first aspect, extracting the track inside along point cloud based on the track face point cloud comprises:

[0021] Extracting left track inside along point cloud in a first predetermined range to the right of the left track face point cloud;

[0022] Extracting right track inside along point cloud in a second predetermined range to the left of the right track face point cloud.

[0023] In a fourth possible implementation manner of the first aspect, in combination with the first aspect, the track face point cloud comprises left track face point cloud and right track face point cloud;

[0024] The fitting according to the track face point cloud to obtain the track face normal vector comprises:

[0025] merge the left rail surface point cloud and the right rail surface point cloud to obtain a rail surface point cloud;

[0026] perform plane fitting on the rail surface point cloud to obtain a rail plane, and obtain a rail surface normal vector according to the rail plane.

[0027] In a fifth possible implementation manner of the first aspect, the determining of the rotation matrix of the point cloud collection device coordinate system to the vehicle body coordinate system according to the rail surface normal vector and the rail line vector comprises:

[0028] perform normalization processing on the rail surface normal vector to obtain a first vector;

[0029] perform normalization processing on the rail line vector to obtain a second vector;

[0030] perform normalization processing on a cross product of the first vector and the second vector to obtain a third vector;

[0031] construct the rotation matrix according to the first vector, the second vector and the third vector.

[0032] In a sixth possible implementation manner of the first aspect, the translation vector comprises an X component, a Y component and a Z component.

[0033] The determining of the translation vector of the point cloud collection device coordinate system to the vehicle body coordinate system according to the rotation matrix and the in-rail point cloud comprises:

[0034] transform the in-rail point cloud according to the rotation matrix to obtain a transformed point cloud, and transform the rail line vector according to the rotation matrix to obtain a transformed rail line vector;

[0035] determine a projection position of a center point of the transformed point cloud on a second plane according to the transformed point cloud and the transformed rail line vector, the second plane being a plane perpendicular to the lane line and including an origin of the vehicle body coordinate system;

[0036] determine the translation vector according to a deviation of the projection position and a center point position of the vehicle body coordinate system, in combination with a train design parameter.

[0037] In a seventh possible implementation manner of the first aspect, in combination with the sixth possible implementation manner of the first aspect, the determining of the projection position of the center point of the transformed point cloud on the second plane according to the transformed point cloud and the transformed rail line vector comprises:

[0038] determine a straight line equation of the left and right rail lines according to the transformed point cloud and the transformed rail line vector;

[0039] determining an intersection point of the straight line equation of the left and right track lines of the point and a second plane, the second plane being a YOZ plane of a vehicle body coordinate system, the vehicle body coordinate system also following a right-hand rule, and a center O of the vehicle body coordinate system being a center of a train head, and an X axis of the vehicle body coordinate system being a train movement direction;

[0040] determining a projection position of the center point of the transformed point cloud collection device on the second plane according to a center position of the intersection line.

[0041] In a possible implementation of the first aspect, the translation vector includes an X component, a Y component, and a Z component.

[0042] According to a deviation of the projection position and a center position of the vehicle body coordinate system, and in combination with train design parameters, the translation vector is determined, including:

[0043] According to an offset amount of the center position of the intersection line and a Y axis of the vehicle body coordinate system, the Z component of the translation vector is determined, and according to an offset amount of the center position of the intersection line and a Z axis of the vehicle body coordinate system, the Y component of the translation vector is determined.

[0044] According to train design parameters, the X component of the translation vector is obtained.

[0045] A second aspect of the embodiment of the present application provides a vehicle body calibration device of a rail transit vehicle, the device including:

[0046] a point cloud extraction unit configured to obtain point cloud data of a track environment in front of the train, and extract track surface point cloud and track inside point cloud according to the point cloud data;

[0047] a point cloud fitting unit configured to fit the track surface point cloud to obtain a track surface normal vector, and fit the track inside point cloud to obtain a track line vector;

[0048] a rotation matrix determination unit configured to determine a rotation matrix of a point cloud collection device coordinate system to a vehicle body coordinate system according to the track surface normal vector and the track line vector;

[0049] a translation vector determination unit configured to determine a translation vector of the point cloud collection device coordinate system to the vehicle body coordinate system according to the rotation matrix and the track inside point cloud.

[0050] A third aspect of the embodiment of the present application provides a vehicle body calibration device of a rail transit vehicle, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the vehicle body calibration device of the rail transit vehicle implements the method according to any one of the first aspect.

[0051] In a fourth aspect, the embodiments of the present application provide a computer program product, which, when executed on a computer, causes the computer to perform the method of the first aspect or any implementation thereof.

[0052] In a fifth aspect, the embodiments of the present application provide a computer-readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the steps of the method of any one of the first aspect.

[0053] In a sixth aspect, the embodiments of the present application provide a chip for implementing the method of any implementation of the first aspect. Specifically, the chip comprises a processor configured to call and execute a computer program from a memory, so that a device installed with the chip performs the method of the first aspect or any implementation thereof.

[0054] Compared with the prior art, the embodiments of the present application have the beneficial effects that: the embodiments of the present application obtain point cloud data of a track environment in front of a train, extract track surface point cloud and track inside point cloud, fit a track surface normal vector from the track surface point cloud, fit a track line vector from the track inside point cloud, determine a rotation matrix of a point cloud collection device coordinate system to a vehicle body coordinate system according to the track surface normal vector and the track line vector, determine a translation vector of the point cloud collection device coordinate system to the vehicle body coordinate system according to the rotation matrix and the track inside point cloud, and complete calibration of the vehicle body coordinate system according to the rotation matrix and the translation vector. This method does not need to build a special calibration site, has low calibration cost, and only needs to collect point cloud data of a track environment in front of a train to automatically complete calibration, which is beneficial to improving calibration efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0056] Figure 1 is an implementation flow diagram of a vehicle body calibration method of a rail transit vehicle provided by the embodiments of the present application;

[0057] Figure 2 is a coordinate definition diagram provided by the embodiments of the present application;

[0058] Figure 3 is a track surface point cloud extraction diagram provided by the embodiments of the present application;

[0059] Figure 4is a translation vector determination schematic diagram provided by an embodiment of the present application;

[0060] Figure 5 is a vehicle body calibration device schematic diagram of a rail transit vehicle provided by an embodiment of the present application;

[0061] Figure 6 is a vehicle body calibration device schematic diagram of a rail transit vehicle provided by an embodiment of the present application. DETAILED DESCRIPTION

[0062] In the following description, for the purpose of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

[0063] In order to illustrate the technical solutions described in the present application, the following will be described through specific embodiments.

[0064] Laser radar has become the core perception unit of the train automatic driving system due to its centimeter-level three-dimensional space measurement capability. By installing the laser radar on the front part of the vehicle body, the system can obtain high-precision point cloud data in front of the train in real time, and construct a dynamic environment model to realize autonomous perception. It should be noted that the reliability of the perception system directly depends on the accuracy of the vehicle body calibration parameters, that is, the spatial transformation matrix of the laser radar coordinate system and the vehicle body coordinate system needs to be accurately solved through mathematical modeling.

[0065] The existing calibration scheme generally has the defects of environmental dependence and artificial dependence: the mainstream method needs to rely on specific static scenes such as standard vertical planes, or rely on artificial arrangement of special calibration objects. Due to the super large physical size of the rail transit equipment and the closed operation scene, the infrastructure investment of the special calibration site has economic bottlenecks; and the method based on special calibration objects needs human participation, which affects the normal operation of the train and has low execution efficiency.

[0066] To solve the above problems, an embodiment of the present application provides a vehicle body calibration method for a rail transit vehicle, as shown in Figure 1 The method comprises the following steps.

[0067] In S101, the point cloud data of the track environment in front of the train is obtained by the point cloud acquisition device, and the track surface point cloud and the track inner edge point cloud are extracted according to the point cloud data.

[0068] The point cloud acquisition device in the embodiment of the present application can include a laser radar, a microwave radar, a depth camera or other devices that can acquire point cloud data.

[0069] The vehicle calibration in this embodiment aims to obtain the relative pose relationship between the vehicle coordinates and the coordinate system of the point cloud acquisition device. For example... Figure 2 In the coordinate system diagram shown for vehicle body calibration, Figure 2 In the left image, O-XYZ represents the coordinate system of the point cloud acquisition device mounted on the roof, and O'-X'Y'Z' represents the vehicle body coordinate system. O' is located on the track plane, and O'X' is parallel to the track centerline. Figure 2 The dashed lines in the right figure coincide, O'Y' is perpendicular to the track line and lies on the track plane, and O'Z' is perpendicular to the track plane. It can be understood that due to vehicle body limitations and installation errors of the point cloud acquisition device, the OX coordinate system of the point cloud acquisition device is not necessarily parallel to O'X' in the vehicle body coordinate system, and the OZ coordinate system of the point cloud acquisition device is not necessarily parallel to O'Z' in the vehicle body coordinate system. Therefore, the goal of vehicle body calibration is to find the transformation relationship between O-XYZ and O'-X'Y'Z' (rotation matrix R, translation vector T).

[0070] It should be noted that the above definition is not unique. In practical applications, O-XYZ varies depending on the installation position of the point cloud acquisition device, and O'-X'Y'Z' can also be adjusted according to application requirements. For example, O' is defined as the position where the first pair of wheels on the vehicle body contacts the track.

[0071] When extracting track surface point clouds and track interior point clouds from point cloud data, the region of interest (ROI) can be determined based on the track width and train height. For example, with a track width of 1.435m and a vehicle height of 3.5m, considering equipment installation deviations, the ROI could be X:[0m,20m], Y:[-1m,1m], Z:[-4m,0m]. The values ​​describing this region are coordinates in the coordinate system of the point cloud acquisition device. This region delineation can exclude irrelevant point clouds far from the track (such as vegetation and buildings) while ensuring complete capture of track surface and interior features.

[0072] To improve processing efficiency and flexibly determine resolution, this embodiment of the application can divide the point cloud within the region of interest into a grid. The grid size is set to 0.1m, which can be adjusted according to actual conditions. Taking the center of the track as the boundary, all grids within the sensing region of interest are divided into a left track grid group and a right track grid group. The point cloud falling within the left track grid group is the left track point cloud, and the point cloud falling within the right track grid group is the right track point cloud.

[0073] The left track point cloud and the right track point cloud are projected onto a first plane, i.e., the XOZ plane, to obtain a first plane projected point cloud distribution. Since there is no occlusion above the track surface in the region of interest, the point cloud projected onto the XOZ plane is viewed from top to bottom in the reverse direction along the Z axis, and a first high-density distribution is formed at the track surface position. Accordingly, the left and right track surfaces can be determined. For example, the density of the first plane projected point cloud distribution can be determined, and the density of the point cloud distribution is compared with a predetermined density threshold to determine a region greater than the density threshold as a first projection region. The left track surface point cloud and the right track surface point cloud projected onto the first projection region can be determined based on the first projection region.

[0074] Considering that the track surface cannot be guaranteed to be horizontal in an actual scene, the XOZ plane projection analysis described above can be performed on point clouds of a certain unit (e.g., one grid size) along the X axis direction each time to obtain the left and right track surface point clouds of the X axis direction slice (e.g., as shown in FIG. 6). Figure 3 The final combined results of all slices are the complete left and right track surface point clouds.

[0075] According to the adjacent relationship between the track surface point cloud and the track inner edge point cloud in the track point cloud data, the left track inner edge point cloud can be extracted in a first predetermined range to the right of the left track surface point cloud, and the right track inner edge point cloud can be extracted in a second predetermined range to the left of the right track surface point cloud based on the determined left track surface point cloud and right track surface point cloud.

[0076] The first predetermined range and the second predetermined range in the present application are generally the same, and can be determined according to the width of the track surface. Typically, 1 / 2 or 1 / 3 of the width of the track surface can be taken.

[0077] In possible implementation manners, the track surface point cloud and the track inner edge point cloud in the point cloud data can also be extracted by a deep learning model.

[0078] In S102, the track surface normal vector is obtained by fitting according to the track surface point cloud, and the track line vector is obtained by fitting according to the track inner edge point cloud.

[0079] The track surface point cloud in the present application includes the left track surface point cloud and the right track surface point cloud. In order to obtain an accurate track surface normal vector, the left track surface point cloud and the right track surface point cloud can be combined to obtain a track surface point cloud, and then the track surface point cloud is fitted to obtain a track plane. The normal vector of the track plane is the track surface normal vector.

[0080] The normal vector of the track plane is a vector perpendicular to the track plane, which describes the inclination direction and angle of the track. By determining the track surface normal vector, the geometric characteristics of the track can be further analyzed to provide important parameters for subsequent coordinate conversion and calibration.

[0081] The process of fitting the track plane can be achieved through mathematical methods such as least squares. By minimizing the error between the point cloud data and the fitted plane, the best plane model is found. During the fitting process, a certain tolerance range can be set to exclude noise points and outliers, improving the accuracy and stability of the fitting. When fitting the track plane point cloud, the distribution density and uniformity of the point cloud data need to be considered. If the point cloud data is not uniformly distributed, it may lead to inaccurate fitting results. Therefore, before performing plane fitting, the point cloud data can be preprocessed, such as removing outliers, smoothing, etc., to improve the accuracy of the fitting.

[0082] Based on the determined track inner side point cloud, linear fitting can be performed on the left track inner side point cloud and the right track inner side point cloud respectively to obtain track line vectors, including left track line vectors and right track line vectors.

[0083] In S103, a rotation matrix from the point cloud acquisition device coordinate system to the vehicle body coordinate system is determined according to the track plane normal vector and the track line vector.

[0084] According to the track plane normal vector and the track line vector, the rotation matrix from the point cloud acquisition device coordinate system to the vehicle body coordinate system can be determined. First, the track plane normal vector is normalized to obtain a first vector, and the track line vector is normalized to obtain a second vector. The track line vector can be a left track line vector or a right track line vector, or the average of the left track line vector and the right track line vector. The cross product of the first vector and the second vector is normalized to obtain a third vector. The rotation matrix is constructed according to the first vector, the second vector and the third vector, which can be expressed as:

[0085] R = [norm(n2) norm(n1 x n2) norm(n1)] -1

[0086] where norm(.) represents the normalization of the vector to ensure that its modulus is 1; n1 x n2 represents the cross product of two vectors, n1 represents the track plane normal vector, and n2 represents the left track line vector. It can be understood that the left track line vector n2 in the above formula can be replaced by the right track line vector n3, or the average of n2 and n3.

[0087] By normalizing the track plane normal vector, a first vector is obtained, which can scale the length of the vector to a unit length to eliminate the influence of the length of the vector on subsequent calculations. For example, assuming that the track plane normal vector is (3, 4, 0) and its length is 5, the first vector obtained after normalization is (0.6, 0.8, 0). Normalization can ensure that the vector has a uniform scale in subsequent calculations, avoiding errors caused by different vector lengths.

[0088] The track line vector represents the direction of the track, and the normalization is also to scale the length to the unit length to obtain a second vector. For example, the track line vector is (1, 2, 2), the length is 3, and the second vector obtained after normalization is (0.333, 0.666, 0.666). The second vector after normalization can accurately represent the direction of the track, and is not affected by the length of the vector.

[0089] The cross product of the first vector and the second vector is normalized to obtain a third vector. The cross product operation is used to generate a new vector perpendicular to the first vector and the second vector, and the normalization ensures that the length of the vector is 1. For example, the first vector is (0.6, 0.8, 0), the second vector is (0.333, 0.666, 0.666), the cross product result is (0.533, -0.4, 0.133), and the third vector obtained after normalization is (0.8, -0.6, 0.2). Obviously, the matrix constructed according to the first vector, the second vector and the third vector meets the basic requirement of the rotation matrix, that is, the unit orthogonal matrix. The construction of the rotation matrix can accurately describe the rotation relationship between the point cloud collection device coordinate system and the vehicle body coordinate system, and provide a basis for subsequent coordinate translation conversion.

[0090] In S104, a translation vector of the point cloud collection device coordinate system to the vehicle body coordinate system is determined according to the rotation matrix and the track inside point cloud.

[0091] After the rotation matrix is determined, the left track inside point cloud, the right track inside point cloud, the left track line vector and the right track line vector can be corrected by rotating the rotation matrix to obtain the rotated point cloud and the line vector. The conversion formula can be expressed as:

[0092] P1'=RP1

[0093] P2'=RP2

[0094] n2'=Rn2

[0095] n3'=Rn3

[0096] Wherein, P1 is the left track inside point cloud, P2 is the right track inside point cloud, n2 is the left track line vector, n3 is the right track line vector, R is the rotation matrix, P1' is the transformed left track inside point cloud, P2' is the transformed right track inside point cloud, n2' is the transformed left track line vector, and n3' is the transformed right track line vector.

[0097] The projection position of the transformed point cloud collection device center point on the second plane can be determined based on the transformed point cloud and the transformed track line vector, the second plane being a plane perpendicular to the lane line and including the origin of the vehicle body coordinate system, and a translation vector is determined according to the deviation of the projection position and the center position of the vehicle body coordinate system, in combination with the train design parameters. The transformed point cloud and the transformed track line vector can determine the track line straight line equation in the three-dimensional space, including the left and right track line straight line equations. The second plane is determined according to the YOZ plane of the vehicle body coordinate system, and the left and right track line straight line equations will have two intersection points with the second plane, and the center position of the connecting line of the two intersection points is the projection position of the transformed point cloud collection device center point on the second plane.

[0098] The translation vector includes an X component, a Y component and a Z component, and the translation vector can be determined based on the deviation of the projection position and the center position of the vehicle body coordinate system, in combination with the train design parameters, including determining the Z component of the translation vector according to the offset of the center position of the connecting line and the Y axis of the vehicle body coordinate system, determining the Y component of the translation vector according to the offset of the center position of the connecting line and the Z axis of the vehicle body coordinate system, and obtaining the X component of the translation vector according to the train design parameters.

[0099] The second plane is the YOZ plane of the vehicle body coordinate system, and the definition of the vehicle body coordinate system also follows the right-hand rule, and the center O of the coordinate system can be set as the center of the vehicle head as shown in Figure 2 The X axis of the vehicle body coordinate system is the direction of train movement.

[0100] The left and right track line equations can be determined based on the converted left and right track line vectors, in combination with the transformed point cloud (transformed track inside point cloud), and the intersection points of the transformed left and right track lines and the second plane can be calculated in combination with the plane equation of the second plane. As shown in Figure 4 The intersection points of the transformed left and right track lines and the second plane are q1' and q2' respectively.

[0101] The Z component of the translation vector can be determined according to the offset of the center position of the connecting line and the Y axis of the vehicle body coordinate system, and the Y component of the translation vector can be determined according to the offset of the center position of the connecting line and the Z axis of the vehicle body coordinate system.

[0102] As shown in Figure 4 The offset of the center position of the connecting line and the Y axis of the vehicle body coordinate system determines the Z component of the translation vector as dz, and the offset of the center position of the connecting line and the Z axis of the vehicle body coordinate system determines the Y component of the balance vector as dy.

[0103] The X component dx of the translation vector can be obtained according to the train design parameters or manual measurement, and in the example of Figure 2 the distance between the origins O and O' in the X' axis direction is dx.

[0104] The determined translation vector T can be expressed as:

[0105] T = [d x -d y -d z ] T

[0106] After the translation vector and the rotation vector are determined, the body calibration of the rail transit vehicle is completed.

[0107] In addition, in the embodiments of the present application, when the rotation matrix and the translation vector are determined, a target function can also be set to realize the estimation of the rotation matrix and the translation vector by minimizing the deviation of the point cloud collection device coordinate system and the vehicle body coordinate system.

[0108] It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0109] Figure 5 A schematic diagram of a body calibration device of a rail transit vehicle provided by the embodiments of the present application is shown, which comprises:

[0110] The point cloud extraction unit 501 is configured to acquire point cloud data of the track environment in front of the train through the point cloud collection device, and extract track surface point cloud and track inside along point cloud according to the point cloud data.

[0111] The point cloud fitting unit 502 is configured to fit according to the track surface point cloud to obtain a track surface normal vector, and fit according to the track inside along point cloud to obtain a track line vector.

[0112] The rotation matrix determination unit 503 is configured to determine a rotation matrix of the point cloud collection device coordinate system to the vehicle body coordinate system according to the track surface normal vector and the track line vector.

[0113] The translation vector determination unit 504 is configured to determine a translation vector of the point cloud collection device coordinate system to the vehicle body coordinate system according to the rotation matrix and the track inside along point cloud.

[0114] Figure 5 The shown body calibration device of the rail transit vehicle corresponds to the shown body calibration method of the rail transit vehicle. Figure 1 The shown body calibration device of the rail transit vehicle corresponds to the shown body calibration method of the rail transit vehicle.

[0115] Figure 6 A schematic diagram of a body calibration device of a rail transit vehicle provided by the embodiments of the present application is shown. As shown in the figure, Figure 6As shown, the vehicle body calibration device 6 of the rail vehicle of the embodiment includes a processor 60, a memory 61, and a computer program 62 stored in the memory 61 and capable of running on the processor 60, for example, a vehicle body calibration program of a rail vehicle. The processor 60 implements the steps in each of the above vehicle body calibration method embodiments of a rail vehicle when executing the computer program 62. Alternatively, the processor 60 implements the functions of each module / unit in each of the above device embodiments when executing the computer program 62.

[0116] For example, the computer program 62 can be divided into one or more modules / units stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program 62 in the vehicle body calibration device 6 of the rail vehicle.

[0117] The vehicle body calibration device 6 of the rail vehicle can be a desktop computer, a notebook, a palm computer, a cloud server, and other computing devices. The vehicle body calibration device of the rail vehicle can include, but is not limited to, a processor 60, a memory 61. Those skilled in the art can understand that Figure 6 The vehicle body calibration device 6 of the rail vehicle is only an example and does not constitute a limitation on the vehicle body calibration device 6 of the rail vehicle, and can include more or fewer components than those shown, or combine certain components, or different components, for example, the vehicle body calibration device of the rail vehicle can also include an input / output device, a network access device, a bus, etc.

[0118] The processor 60 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0119] The memory 61 can be an internal storage unit of the rail vehicle body calibration device 6, such as a hard disk or a memory of the rail vehicle body calibration device 6. The memory 61 can also be an external storage device of the rail vehicle body calibration device 6, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the rail vehicle body calibration device 6. Further, the memory 61 can also include both the internal storage unit and the external storage device of the rail vehicle body calibration device 6. The memory 61 is used to store the computer program and other programs and data required by the rail vehicle body calibration device. The memory 61 can also be used to temporarily store data that has been output or will be output.

[0120] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for easy distinction, and do not limit the protection scope of the application. The specific working process of the units and modules in the system can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0121] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0122] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the application.

[0123] In the embodiments of the present application, it should be understood that the disclosed device and method can be implemented in other manners. For example, the described device embodiments are merely schematic. Taking the division of the modules or units as an example, the division can be different, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0124] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0125] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware, or in the form of a software functional unit.

[0126] The integrated module / unit, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such an understanding, all or part of the flow of the above-mentioned embodiment methods can be completed by computer program instruction-related hardware. The computer program can be stored in a computer readable storage medium. When the processor executes the computer program, the steps of each method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable file or some intermediate form. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0127] In addition, the embodiments of the present application also provide a computer program product, which, when running on a computer, causes the computer to execute the method in each of the above implementations.

[0128] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method of calibrating a body of a rail vehicle, characterized in that, The method comprises: acquiring point cloud data of a track environment in front of a train, and extracting track surface point cloud and track inside edge point cloud according to the point cloud data; fitting according to the track surface point cloud to obtain a track surface normal vector, and fitting according to the track inside edge point cloud to obtain a track line vector; determining a rotation matrix from a point cloud acquisition device coordinate system to a vehicle body coordinate system according to the track surface normal vector and the track line vector; determining a translation vector from the point cloud acquisition device coordinate system to the vehicle body coordinate system according to the rotation matrix and the track inside edge point cloud.

2. The method of claim 1, wherein, The extraction of the track surface point cloud and the track inside edge point cloud according to the point cloud data comprises: determining a region of interest according to a track width and a train height; performing point cloud division on point cloud data in the region of interest to obtain left track point cloud and right track point cloud; projecting the left track point cloud and the right track point cloud onto a first plane to obtain a first plane projected point cloud distribution, the first plane being an XOZ plane of a point cloud acquisition device coordinate system, the center O of the point cloud acquisition device coordinate system being a center position of the point cloud acquisition device, the point cloud acquisition device coordinate system complying with the right-hand rule, with the X axis being defined as a direction from the center of the point cloud acquisition device coordinate system outward and perpendicular to the emission plane, the right-hand index finger direction being the X axis direction, the middle finger direction being the Y axis, and the thumb direction being the Z axis; extracting the track surface point cloud according to the first plane projected point cloud distribution, and extracting the track inside edge point cloud based on the track surface point cloud.

3. The method of claim 2, wherein, The track surface point cloud comprises left track surface point cloud and right track surface point cloud; The extraction of the track surface point cloud according to the first plane projected point cloud distribution comprises: analyzing the first plane projected point cloud distribution from top to bottom along the reverse direction of the Z axis to determine the first plane projected point cloud distribution density, and determining a first projection region with a point cloud distribution density greater than a predetermined density threshold; extracting left track surface point cloud and right track surface point cloud projected onto the first projection region.

4. The method of claim 3, wherein, The extraction of the track inside edge point cloud based on the track surface point cloud comprises: extracting left track inside edge point cloud in a first predetermined range to the right of the left track surface point cloud; extracting right track inside edge point cloud in a second predetermined range to the left of the right track surface point cloud.

5. The method of claim 1, wherein, The track surface point cloud comprises left track surface point cloud and right track surface point cloud; The fitting according to the track surface point cloud to obtain a track surface normal vector comprises: fitting the left track surface point cloud and the right track surface point cloud to obtain track surface point cloud; performing plane fitting on the track surface point cloud to obtain a track plane, and obtaining a track surface normal vector according to the track plane.

6. The method of claim 1, wherein, The determination of a rotation matrix from a point cloud acquisition device coordinate system to a vehicle body coordinate system according to the track surface normal vector and the track line vector comprises: performing normalization processing on the track surface normal vector to obtain a first vector; performing normalization processing on the track line vector to obtain a second vector; performing normalization processing on the cross product of the first vector and the second vector to obtain a third vector; constructing the rotation matrix according to the first vector, the second vector, and the third vector.

7. The method of claim 1, wherein, The translation vector of the point cloud acquisition device coordinate system to the vehicle body coordinate system is determined according to the rotation matrix and the in-track point cloud, and the translation vector includes an X component, a Y component, and a Z component. The in-track point cloud is transformed according to the rotation matrix to obtain a transformed point cloud, and the track line vector is transformed according to the rotation matrix to obtain a transformed track line vector. A projection position of the center point of the transformed point cloud on a second plane is determined according to the transformed point cloud and the transformed track line vector, and the second plane is a plane that is perpendicular to the lane line and includes the origin of the vehicle body coordinate system. The translation vector is determined according to the deviation of the projection position and the position of the center point of the vehicle body coordinate system, and in combination with the train design parameters.

8. The method of claim 7, wherein, The projection position of the center point of the transformed point cloud on the second plane is determined according to the transformed point cloud and the transformed track line vector, and the method includes the following steps. The left and right track line straight line equations are determined according to the transformed point cloud and the transformed track line vector. The intersection points of the left and right track line straight line equations and the second plane are determined. The projection position of the center point of the transformed point cloud on the second plane is determined according to the center position of the intersection line.

9. The method of claim 7, wherein, The translation vector includes an X component, a Y component, and a Z component. The translation vector is determined according to the deviation of the projection position and the position of the center point of the vehicle body coordinate system, and in combination with the train design parameters, and the method includes the following steps. The Z component of the translation vector is determined according to the offset of the center position of the intersection line and the Y axis of the vehicle body coordinate system, and the Y component of the translation vector is determined according to the offset of the center position of the intersection line and the Z axis of the vehicle body coordinate system. The X component of the translation vector is obtained according to the train design parameters.

10. A vehicle body calibration device for a rail vehicle, characterized by comprising: The device includes: A point cloud extraction unit is configured to acquire point cloud data of a track environment in front of a train through a point cloud acquisition device, and extract track surface point cloud and in-track point cloud according to the point cloud data. A point cloud fitting unit is configured to fit the track surface point cloud to obtain a track surface normal vector, and fit the in-track point cloud to obtain a track line vector. A rotation matrix determination unit is configured to determine a rotation matrix of a point cloud acquisition device coordinate system to a vehicle body coordinate system according to the track surface normal vector and the track line vector. A translation vector determination unit is configured to determine a translation vector of the point cloud acquisition device coordinate system to the vehicle body coordinate system according to the rotation matrix and the in-track point cloud.

11. A rail vehicle body calibration apparatus comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein, The processor executes the computer program, so that the vehicle body calibration device of the rail vehicle implements the method according to any one of claims 1-9.

12. A computer program product comprising computer program instructions, characterised in that, The computer program is executed, so that the method according to any one of claims 1-9 is executed.

Images