Railway turnout heart measurement method and system based on airborne LiDAR point cloud

Abstract

This invention discloses a method and system for measuring the centerline of a railway turnout based on airborne LiDAR point clouds. The method utilizes an airborne LiDAR device to scan the target railway turnout and its surrounding area, and preprocesses the data to obtain a high-precision point cloud. Straight and curved rail segments are selected within the turnout area, and the centerline coordinates of these rails are extracted. The centerlines of the straight and curved rails are then fitted to obtain the centerlines of the straight and curved rails. The virtual intersection point of the straight line between the centerline of the curved rail and the centerline of the straight rail is obtained to determine the coordinates of the turnout centerline. This method can automatically and accurately extract the rail centerlines from the three-dimensional point cloud of the turnout, and reliably calculate the three-dimensional coordinates of the theoretical turnout centerline by fitting the straight line and calculating the intersection point. It has the advantages of high efficiency, high accuracy, and strong anti-interference capability.

Description

Technical Field

[0001] This invention relates to the field of railway turnout core measurement and detection technology, specifically to a railway turnout core measurement method and system based on airborne LiDAR point clouds. Background Technology

[0002] Railway turnouts are a key component of the track structure, and accurate measurement of their geometry is crucial for ensuring safe train operation. The turnout point (also known as the theoretical tip of the frog) is the core geometric point of the turnout, and accurate determination of its coordinates is the foundation for turnout positioning, geometric dimension inspection, and maintenance.

[0003] Currently, fork point measurement mainly relies on traditional manual measurement methods:

[0004] 1. Total Station Survey: This method requires surveyors to be on the track and use a prism to conduct multi-point measurements, indirectly calculating the coordinates of the branch center through methods such as resection. This method is inefficient, greatly affected by weather and lighting conditions, and the measurement time within the track maintenance window is limited, posing safety hazards.

[0005] 2. Track gauge measurement with drawstring: This method is more primitive, has low accuracy, and relies entirely on the operator's experience, resulting in poor reliability.

[0006] These traditional methods all suffer from significant drawbacks, including low measurement efficiency, insufficient automation, susceptibility to subjective errors, and substantial interference with train operation safety during track measurements. In recent years, 3D laser scanning technology, due to its non-contact, high-density, and high-precision characteristics, has been introduced into the railway surveying field. However, existing point cloud processing technologies are mostly focused on rail extraction and gauge detection. For the abstract theoretical intersection point of a turnout, there is still a lack of an effective method to automatically and accurately calculate its 3D coordinates directly from disordered, massive point cloud data. Currently, it is usually necessary to rely on manual identification of features such as the stock rail and switch rail in the point cloud for rough estimation, resulting in low automation and difficulty in guaranteeing accuracy. Therefore, there is an urgent need for a method that can automatically, efficiently, and accurately determine the location of railway turnout points based on dense point cloud data. Summary of the Invention

[0007] This invention provides a method and system for measuring the center of a railway turnout based on airborne LiDAR point clouds, in order to solve the problems of low measurement efficiency, insufficient automation, susceptibility to subjective errors, and significant interference with train operation safety caused by on-track measurement in the existing technology.

[0008] According to the first aspect, one embodiment provides a method for measuring the center of a railway turnout based on airborne LiDAR point clouds, the method comprising:

[0009] The target railway turnout and surrounding area were scanned using an airborne LiDAR device to obtain high-density raw 3D point cloud data.

[0010] The original 3D point cloud data obtained from the scan is preprocessed to obtain high-precision point cloud results.

[0011] Based on the high-precision point cloud results, according to the spatial geometric characteristics of the railway turnout structure, straight and curved main rail sections in the turnout area are selected, and the centerline coordinate points of the straight and curved main rails are extracted.

[0012] Based on the coordinates of the centerline of the straight track, the centerline of the straight track is obtained by fitting according to the principle of minimizing track shifting; based on the coordinates of the centerline of the curved track, the curve segment is automatically fitted and optimized with reference to the curve parameters in the ledger, and the centerline of the curved track is obtained by fitting.

[0013] The virtual intersection of the centerline of the curve basic rail and the centerline of the straight basic rail is obtained to get the coordinates of the turnout center position.

[0014] Calculate the distance from the tip of the turnout switch rail to the center of the turnout, and estimate the center of the turnout mileage based on the mileage information of the tip of the turnout switch rail in the ledger, thus obtaining the center of the turnout measurement results.

[0015] Furthermore, airborne LiDAR equipment is used to scan the target railway turnout and surrounding area to obtain high-density raw 3D point cloud data, specifically including:

[0016] Base station control networks and target control networks are pre-deployed along the railway line; the base station control network is used for the construction of UAV flight base stations to meet the requirements of subsequent PPK calculation; the target control network is used for point cloud correction to improve the absolute accuracy of point cloud results.

[0017] The base station control network prioritizes the use of the existing control network. When the location and accuracy of the existing control network do not meet the requirements for base station construction, a separate base station control network will be deployed.

[0018] The target control network is set up manually, along both sides of the railway land boundary. The point markers are prefabricated target plates. The three-dimensional coordinates of the target control points are measured using GNSS rapid static and leveling methods.

[0019] When drones fly, the flight path is laid along both sides of the railway line to avoid entering the railway land boundary, and the flight path overlap and radar parameters are set reasonably.

[0020] Furthermore, the raw 3D point cloud data obtained from the scan is preprocessed to obtain high-precision point cloud results, specifically including:

[0021] The initial trajectory line file is obtained by fusing and solving airborne GNSS data, inertial navigation data, and ground base station GNSS data using integrated navigation solution software.

[0022] The initial trajectory file is corrected by using target control points deployed along the line to obtain a high-precision trajectory line that is consistent with the coordinates of the target control network.

[0023] The corrected high-precision trajectory line is fused with the scanned data again to obtain the point cloud results in the WGS84 coordinate system.

[0024] The point cloud results in the WGS84 coordinate system are transformed into the engineering independent coordinate system to obtain high-precision point cloud results.

[0025] Furthermore, based on the high-precision point cloud results and according to the spatial geometric characteristics of the railway turnout structure, straight and curved main rail sections in the turnout area are selected, and the centerline coordinates of the straight and curved main rails are extracted, specifically including:

[0026] For straight and curved basic track segments, the methods for extracting the coordinate points of the track centerline include:

[0027] Based on the high-precision point cloud results, combined with the pre-calibrated structural parameters of the equipment, the laser point cloud on the track surface is automatically classified by the algorithm.

[0028] Based on the classified track surface laser point cloud, POS position and attitude information and equipment calibration parameters, a local coordinate system of the cross-section track surface center at any mileage is constructed. Under the corresponding local coordinate system, the laser point cloud of the rail top area is filtered according to a preset spatial range to obtain the rail head point data of any cross section.

[0029] Select equally spaced cross-sectional rail head point cloud data and register them with the preset standard rail head model. Then, determine the left and right rail vertices of each cross section using the registered and fitted rail head model.

[0030] The average plane coordinates of the left and right rail vertices are taken as the plane coordinates of the centerline point of the track, and the minimum elevation of the left and right rail vertices is taken as the elevation of the centerline point of the track.

[0031] Furthermore, the formulas for calculating the plane coordinates and elevation of the centerline points on the track are as follows:

[0032]

[0033]

[0034]

[0035] in, , , These are the plane X-coordinate, plane Y-coordinate, and elevation of the left rail vertex, respectively. , , These are the plane X-coordinate, plane Y-coordinate, and elevation of the right-side vertex, respectively. , , These are the plane X-coordinate, plane Y-coordinate, and elevation of the centerline point of the track.

[0036] Furthermore, based on the coordinates of the centerline of the basic straight track, and following the principle of minimizing track shifting, the centerline of the basic straight track is fitted, specifically including:

[0037] The orthogonal least squares fitting algorithm is used for line fitting. The relevant parameters of the line equation are solved based on the criterion of minimizing the sum of squared residuals of the orthogonal distances from the measurement points to the fitted line.

[0038] Furthermore, based on the coordinates of the basic centerline of the curve, the curve segment is automatically fitted and optimized with reference to the parameters of the ledger curve to obtain the basic centerline of the curve, specifically including:

[0039] When fitting a curve, the curve is first roughly segmented. The basic curve track consists of "straight line-curve-straight line-curve-straight line". The slope of each two adjacent coordinate points of the centerline of the basic curve track is calculated. Based on the change in slope, the boundary between the straight line segment and the curve segment of the centerline of the basic curve track is roughly distinguished.

[0040] For approximate straight line segments, an orthogonal least squares fitting algorithm is used to fit the straight line; then, circular curves and transition curves are fitted. After the fitting is completed, the coordinates of the straight / transition / transition point and the circular / transition point are calculated; the calculation is repeated multiple times until the coordinate error is lower than the preset threshold, at which point the iteration stops.

[0041] Furthermore, the virtual intersection point of the straight line between the centerline of the curved track and the centerline of the straight track is obtained to determine the coordinates of the turnout center position, specifically including:

[0042] Connect the transition point and the straight-transition point obtained after fitting the centerline of the basic curve track, and then extend it towards the center of the fork, intersecting with the centerline of the basic straight track obtained by fitting. The plane coordinates of the center of the fork are the coordinates of the intersection point. The elevation of the center of the fork is obtained by interpolating the elevations of the two adjacent coordinate points before and after the intersection point on the centerline of the basic straight track.

[0043] Furthermore, the distance from the tip of the turnout switch rail to the turnout center position is calculated. Based on the turnout switch rail tip mileage information in the ledger, the turnout center mileage is deduced, and the turnout center measurement results are obtained, specifically including:

[0044] The position of the turnout switch tip is manually identified from the point cloud and projected onto the fitted basic rail centerline. Then, the distance between the projection point and the turnout center is calculated, and the mileage of the turnout switch tip is checked in the track maintenance log to deduce the mileage information of the turnout center.

[0045] According to a second aspect, one embodiment provides a railway turnout center measurement system based on airborne LiDAR point clouds, the system comprising:

[0046] The airborne point cloud data acquisition module is used to scan the target railway turnout and surrounding area using airborne LiDAR equipment to acquire high-density raw three-dimensional point cloud data.

[0047] The airborne point cloud data preprocessing module is used to preprocess the raw 3D point cloud data obtained by scanning to obtain high-precision point cloud results.

[0048] The turnout basic rail centerline extraction module is used to select straight and curved basic rail segments in the turnout area based on high-precision point cloud results and the spatial geometric characteristics of the railway turnout structure, and extract the centerline coordinate points of the straight and curved basic rails.

[0049] The turnout area centerline fitting module is used to fit the centerline of the straight track based on the coordinate points of the centerline of the straight track, according to the principle of minimizing the track shifting amount; and to automatically fit the centerline of the curve segment based on the coordinate points of the centerline of the curve, and optimize it with reference to the curve parameters in the ledger, so as to fit the centerline of the curve.

[0050] The turnout center coordinate calculation module is used to obtain the virtual intersection point of the straight line between the centerline of the curve basic rail and the centerline of the straight basic rail to obtain the coordinates of the turnout center position.

[0051] The turnout center mileage calculation and output module is used to calculate the distance from the tip of the turnout switch rail to the turnout center position. Based on the turnout switch rail tip mileage information in the ledger, the turnout center mileage is calculated, and the turnout center measurement results are obtained.

[0052] This invention provides a method and system for measuring the center of a railway turnout based on airborne LiDAR point clouds, which has the following advantages:

[0053] 1. The method for measuring turnout points based on airborne LiDAR dense point cloud proposed in this invention can solve the problems of safety hazards and poor accuracy of traditional manual track measurement. It uses a non-contact method of UAV to acquire high-precision point clouds, without the need to apply for "track closure", avoiding manual on-line measurement, reducing the safety risks of field data acquisition, and improving the overall operation efficiency.

[0054] 2. In the calculation of the turnout center, this invention adopts a rail head matching centerline extraction and automatic line shape fitting algorithm. It uses millions to tens of millions of point cloud data to accurately fit the track geometry through the algorithm, eliminating the random errors of traditional manual single-point measurement, and effectively avoiding the problem of turnout center offset caused by the deviation of the basic rail clamp straight line selection. The measurement accuracy can reach the millimeter level.

[0055] 3. The method of this invention can obtain the coordinates of the turnout center and a complete three-dimensional model of the turnout at the same time, which can be used for the detection and analysis of geometric parameters in more dimensions, resulting in richer results. Attached Figure Description

[0056] Figure 1 A flowchart illustrating a method for measuring the center of a railway turnout based on airborne LiDAR point clouds, provided as an embodiment of the present invention;

[0057] Figure 2 A flowchart illustrating a specific implementation of a railway turnout core measurement method based on airborne LiDAR point clouds, as provided in one embodiment of the present invention;

[0058] Figure 3 A schematic diagram of the target pattern and layout in a railway turnout center measurement method based on airborne LiDAR point cloud provided in an embodiment of the present invention;

[0059] Figure 4 A schematic diagram of the current status of the railway turnout area in a railway turnout center measurement method based on airborne LiDAR point cloud provided in an embodiment of the present invention;

[0060] Figure 5 A schematic diagram illustrating the principle of track centerline extraction based on rail head model matching in a railway turnout centerline measurement method based on airborne LiDAR point clouds, provided as an embodiment of the present invention.

[0061] Figure 6 A schematic diagram of the basic track plane alignment of a railway turnout in a railway turnout center measurement method based on airborne LiDAR point cloud provided in an embodiment of the present invention;

[0062] Figure 7 A schematic diagram of a transition curve in a railway turnout center measurement method based on airborne LiDAR point clouds, provided as an embodiment of the present invention;

[0063] Figure 8 This is a schematic diagram illustrating the calculation principle of the railway turnout center in a railway turnout center measurement method based on airborne LiDAR point clouds, provided as an embodiment of the present invention. Detailed Implementation

[0064] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0065] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0066] The first embodiment of this invention provides a method for measuring the center of a railway turnout based on airborne LiDAR point clouds. This method aims to solve the problems of difficulty in manual track inspection and low measurement reliability. It rapidly acquires high-density point cloud data of the railway turnout area through non-contact measurement using an unmanned aerial vehicle (UAV). Algorithms are used to automatically extract the track centerline and automatically fit the track type. The coordinates of the turnout center are obtained using the intersection of straight segments. This method boasts a high degree of automation and reliable measurement accuracy, meeting the needs of station design. The following section combines... Figure 1 and Figure 2 Please provide a detailed explanation.

[0067] like Figure 1 As shown, in step S100, an airborne LiDAR device is used to scan the target railway turnout and surrounding area to obtain high-density original three-dimensional point cloud data.

[0068] In this embodiment, airborne point cloud data acquisition involves the following steps: Before data acquisition, a base station control network and a target control network need to be deployed at the station. The base station control network is used for setting up UAV flight base stations to meet the subsequent PPK calculation requirements. The target control network is used for point cloud correction to improve the absolute accuracy of the point cloud results. Existing control networks can be used as a priority for the base station control network. If the location and accuracy do not meet the requirements for base station setup, a separate base station control network needs to be deployed. The target control network is deployed manually, along both sides of the railway land boundary (protective net). Prefabricated target boards are used as point markers, and the three-dimensional coordinates of the target control points are measured using GNSS rapid static and leveling methods. During UAV flight, the flight path is laid along both sides of the railway line to avoid entering the railway land boundary. The flight path overlap and radar parameters need to be set reasonably to ensure that the point cloud density is better than 2500 points / square meter.

[0069] The specific deployment method for the target control network is as follows:

[0070] Step 101: First, design the target style, such as... Figure 3 As shown, it can be designed as a rectangle (size greater than 50cm), made of KT board (thickness not less than 5mm), printed with alternating black and white colors, and finally covered with a frosted film to increase the reflectivity of the laser.

[0071] Step 102: Pre-select target control points on the network satellite map and set them up in pairs outside the protective netting along both sides of the line at a spacing of 300 meters. When setting them up on the ground, the angle between the diagonal direction of the target plate and the line should be kept as close as possible to 90° so that the laser scanning direction is perpendicular to the direction of the target reflection stripes. This will increase the number of point clouds on the target and make it easier to identify the target center later.

[0072] Step 103: The target plane is measured using GNSS rapid static measurement, with an observation time of 25-30 minutes, and at least two connections are made with the base station control network; the elevation is measured using leveling forward and backward measurement.

[0073] Step 104: Solve and adjust the target plane and elevation measurement data to obtain the final target control point results.

[0074] The drone flight parameters settings mainly include flight altitude, flight speed, laser emission frequency, scanning linear velocity, radar field of view, and other parameters. In order to accurately identify track information in the turnout area, the point density should be no less than 2,500 points / square meter, and the point cloud thickness should not exceed 2 cm.

[0075] like Figure 1 As shown, in step S200, the original three-dimensional point cloud data obtained by scanning is preprocessed to obtain high-precision point cloud results.

[0076] In this embodiment, the airborne point cloud data preprocessing involves using integrated navigation calculation software to fuse airborne GNSS data, inertial navigation data, and ground base station GNSS data to obtain an initial trajectory file. Then, using target control points deployed along the route, the trajectory file undergoes correction, re-fusion, coordinate system transformation, and noise reduction filtering to ultimately obtain a high-precision point cloud result. Details are as follows:

[0077] Step 201: Use the integrated navigation solution software (Inertial Explorer) to perform fusion calculation of airborne GNSS data, inertial navigation data and ground base station GNSS data. Generally, a tightly coupled method is adopted to obtain the initial trajectory line file.

[0078] Step 202: Using the POS fusion point cloud data from the initial solution, identify the target position from the point cloud. The center position of the target may not be exactly the same as the scanning point. The coordinates of the center point can be obtained by fitting the target contour.

[0079] Step 203: At this point, the target has two sets of coordinates: point cloud coordinates and control network coordinates. The POS trajectory line is corrected by back-correction through parameter calculation to obtain a high-precision POS trajectory line that is consistent with the target control network coordinates.

[0080] Step 204: Use the corrected trajectory line to fuse and solve the scan data again to obtain the point cloud results in the WGS84 coordinate system.

[0081] Step 205: Calculate 7 parameters using the two sets of coordinates (WGS84 geodetic height results and engineering independent coordinate system leveling height results) of the base station control network covering the survey area, and convert the point cloud results to the engineering independent coordinate system to obtain the final point cloud results.

[0082] like Figure 1 As shown, in step S300, based on the high-precision point cloud results and according to the spatial geometric characteristics of the railway turnout structure, the straight main rail and curved main rail segments in the turnout area are selected, and the centerline coordinate points of the straight main rail and curved main rail are extracted.

[0083] In this embodiment, the centerline extraction of the turnout main rail is performed as follows: Based on the spatial geometric features of the railway turnout structure, straight and curved main rail sections are manually selected, and the center coordinates of the rail surface points of the left and right rails are automatically extracted at equal intervals using a software rail head matching algorithm. The three-dimensional coordinates of the centerlines of the straight and curved main rails are then calculated.

[0084] Turnout structure, such as Figure 4 As shown, it is divided into a straight basic track and a curved basic track. The curved basic track is composed of two circular curves, short straight lines, and transition curves in opposite directions. The so-called intersection point is the virtual intersection point between the straight line in the middle of the circular curve track and the centerline of the straight basic track.

[0085] The extraction of the basic rail centerline of a turnout can be achieved through the following steps:

[0086] Step 301: First, use the POS position and attitude information data generated by airborne LiDAR fusion, and then combine it with the structural information parameters calibrated by the device itself to automatically classify the orbital surface point cloud through an algorithm.

[0087] Step 302: Construct a cross-sectional rail surface center coordinate system at any mileage using the classified rail surface laser point cloud, POS attitude information, and equipment calibration parameters. Then, filter the point cloud at the rail top under this rail surface coordinate system according to a spatial dimension of 20cm in plane and 20cm in elevation to finally obtain the rail head point data for any cross-section.

[0088] Step 303: Register the cross-sectional point cloud data with a specified spacing to a standard rail head model (e.g., 60-rail, 50-rail). Figure 5 As shown in the figure, the left and right rail vertices are then obtained using the fitted railhead model.

[0089] Step 304: The plane coordinates of the track centerline are taken as the average of the left and right rail vertices, and the elevation is taken as the minimum of the left and right rail top elevations. Specifically, the following formula can be used for calculation:

[0090] (1)

[0091] (2)

[0092] (3)

[0093] In the formula: , These are the three-dimensional coordinates of the centerline of the track; , These are the three-dimensional coordinates of the left edge vertex; , These are the three-dimensional coordinates of the right-side vertex.

[0094] The centerline of the straight basic rail can be extracted at 5-meter intervals. Since the curved basic rail is very short (usually 10-20 meters) between the straight line and the curve, it needs to be extracted at 2-meter or 1-meter intervals to ensure that there are enough feature points for subsequent fitting.

[0095] like Figure 1 As shown, in step S400, based on the coordinate points of the basic centerline of the straight track, the basic centerline of the straight track is fitted according to the principle of minimizing the track shifting amount; based on the coordinate points of the basic centerline of the curve track, the curve segment is automatically fitted and optimized with reference to the ledger curve parameters to fit the basic centerline of the curve track.

[0096] In this embodiment, the centerline fitting of the turnout area is as follows: based on the extracted coordinates of the straight main rail, the centerline is fitted according to the principle of minimizing the track shifting amount to obtain the fitted straight rail centerline; then, based on the coordinates of the curved rail centerline, the curved segment is automatically fitted and optimized with reference to the ledger curve parameters to obtain the curved rail centerline and the five major pile elements of the curve (straight-to-slow point ZH, slow-to-round point HY, curved center point QZ, round-to-slow point YH, and slow-to-straight point HZ).

[0097] Centerline fitting in the turnout area is divided into straight line fitting and curve fitting, following the principle of minimizing track shifting. The specific implementation includes the following steps:

[0098] Step 401: The orthogonal least squares fitting algorithm is used for line fitting. The relevant parameters of the line equation are solved based on the criterion of minimizing the sum of squared residuals of the orthogonal distances from the measurement points to the fitted line.

[0099] The calculation process for orthogonal least squares linear fitting is as follows:

[0100] Let the equation of the line be:

[0101] (4)

[0102] in The parameters are for fitting the straight line. These are the plane coordinates of the measurement points on the straight section of the track. First, calculate using the coordinates of the two endpoints of this straight line. initial value .

[0103] From the formula for the distance from a point to a line, the sum of the squares of the distances S from any measured point on the straight segment of the track to the fitted line is:

[0104] (5)

[0105] After expanding and linearizing using Taylor's formula, we can obtain... The error equation for the unknown parameters simplifies as follows, where The straight line represents the initial values ​​of the coefficients. Let B be the measured coordinates of the points, and let B be the coefficient matrix. It is a column vector representing the current parameter estimates. Below, the negative value of the square of the distance from each measured point to the line (after normalization). To represent the coefficient correction value:

[0106] (6)

[0107] (7)

[0108] (8)

[0109] The coefficients of the straight line are:

[0110] (9)

[0111] Step 402: When fitting the curve, first perform a rough segmentation. The basic track of a turnout curve is generally composed of "straight line-curve-straight line-curve-straight line", such as... Figure 6 As shown. The slope of every two adjacent coordinate points of the centerline of the curved track is calculated by (Formula 10), and the boundary between the straight segment and the curved segment is roughly distinguished based on the change in slope.

[0112] (10)

[0113] In the formula: m represents the slope between two adjacent measuring points. , Represents the plane coordinates of any midline point.

[0114] Step 403: For the approximate straight line segment, use the orthogonal least squares fitting algorithm to fit the line. Extend the fitted two lines to intersect. At this point, the center of the circle lies on the angle bisector of the interior angle at the intersection of the two lines. Assume the equation of the angle bisector is... Assuming To fit the center coordinates and radius of the circular curve, Let be the plane coordinates of the measured points on the circular curve segment of the track. Then, the distance S from any measured point on the circular curve to the curve can be expressed by the following formula:

[0115] (11)

[0116] After expanding and linearizing using Taylor's formula, we can obtain... The error equation for the unknown parameters simplifies as follows, where We can first use ordinary least squares fitting, with other parameters having the same meaning as the straight line:

[0117] (12)

[0118] (13)

[0119] (14)

[0120] The parameters of the circular curve are:

[0121] (15)

[0122] Necessary conditions are Not infinite or null, and Greater than 0, The distance is less than the distance from the center of the circle to the two intersecting lines. If the approximate segmentation result is too poor or contains gross errors, the necessary condition may not be met. If the necessary condition is not met, iterate through all radii that meet the specifications and take the radius that satisfies the necessary condition and has the smallest cumulative deviation as the result.

[0123] Step 404: After obtaining the fitting parameters for the straight line and circular curve, calculate the transition curve, referring to the following formula:

[0124] (16)

[0125] (17)

[0126] (18)

[0127] in These represent the offset within the circular curve, the length of the transition curve, and the tangent angle of the transition curve, respectively. a, b, and c are parameters of the straight line at the straight end of the circular-to-transition curve. The coordinates of the center and radius of the circular curve at the end of the curve are given.

[0128] Step 405: After fitting the circular curve and the transition curve, calculate the coordinates of the points of the straight / transition curve and the circular / transition curve. The calculation principle is as follows: Figure 7 As shown.

[0129] Straight-slow (gradually straight) point

[0130] (19)

[0131] (20)

[0132] in( ) are the coordinates of the straight and gentle points, ( ( ) represents the coordinates of the intersection point of the circular curves. The coordinate azimuth angle from the intersection point JD (the intersection point is the point where the two straight lines of the standard alignment intersect) to the ZH point can be calculated from the slope of the line. T is the tangent length, calculated using the following formula:

[0133] (twenty one)

[0134] in That is, the length from the center of the circle to the foot of the perpendicular from the line segment to the intersection point. , This is the distance from the center of the circle to the intersection point.

[0135] Slow-moving (slow-moving) point

[0136] (twenty two)

[0137] (twenty three)

[0138] in, Let these be the coordinates of the curve's transition point. The coordinates of point ZH or HZ in the independent coordinate system of the railway line project. The coordinates of point HY or YH in the independent coordinate system of the transition curve. Let ZH be the azimuth angle from point ZH. K is the deflection factor, and the center of the circle is located at the vector. K=1 when on the left and K=-1 when on the right.

[0139] (twenty four)

[0140] (25)

[0141] Step 406: Repeat steps 402-405. Through repeated iterations, stop when the coordinates of the five major piles tend to stabilize (the difference between two coordinates is less than 1cm).

[0142] like Figure 1 As shown, in step S500, the virtual intersection point of the straight line between the centerline of the curved basic rail and the centerline of the straight basic rail is obtained to obtain the coordinates of the turnout center position.

[0143] In this embodiment, as Figure 8 As shown, the calculation of the bifurcation center coordinates involves connecting the transition points and straight-transition points of the fitted curve base rail, and then extending the line towards the bifurcation center to intersect with the straight line fitted to the straight base rail. The plane coordinates of the bifurcation center are the coordinates of the intersection point, and the elevation coordinates are obtained by interpolating the elevations of the coordinate points before and after the straight base rail.

[0144] like Figure 1 As shown, in step S600, the distance from the tip of the turnout switch rail to the center of the turnout is calculated, and the center of the turnout is estimated based on the mileage information of the tip of the turnout switch rail in the ledger, thus obtaining the measurement results of the center of the turnout.

[0145] In this embodiment, the calculation and output of the turnout point mileage are as follows: The turnout point mileage is usually an important piece of information in the measurement results. The position of the tip of the switch rail is manually identified from the point cloud and projected onto the fitted straight line of the main rail. Then, the distance between the point and the turnout point calculated in step S500 is calculated. The mileage of the tip of the switch rail in the track maintenance ledger is consulted to deduce the mileage information of the turnout point. The turnout point measurement results are organized in the format of "mileage + east coordinate + north coordinate + elevation".

[0146] Corresponding to the above-disclosed method for measuring the center of a railway turnout based on airborne LiDAR point clouds, this invention also discloses a system for measuring the center of a railway turnout based on airborne LiDAR point clouds, which specifically includes:

[0147] The airborne point cloud data acquisition module is used to scan the target railway turnout and surrounding area using airborne LiDAR equipment to acquire high-density raw three-dimensional point cloud data.

[0148] The airborne point cloud data preprocessing module is used to preprocess the raw 3D point cloud data obtained by scanning to obtain high-precision point cloud results.

[0149] The turnout basic rail centerline extraction module is used to select straight and curved basic rail segments in the turnout area based on high-precision point cloud results and the spatial geometric characteristics of the railway turnout structure, and extract the centerline coordinate points of the straight and curved basic rails.

[0150] The turnout area centerline fitting module is used to fit the centerline of the straight track based on the coordinate points of the centerline of the straight track, according to the principle of minimizing the track shifting amount; and to automatically fit the centerline of the curve segment based on the coordinate points of the centerline of the curve, and optimize it with reference to the curve parameters in the ledger, so as to fit the centerline of the curve.

[0151] The turnout center coordinate calculation module is used to obtain the virtual intersection point of the straight line between the centerline of the curve basic rail and the centerline of the straight basic rail to obtain the coordinates of the turnout center position.

[0152] The turnout center mileage calculation and output module is used to calculate the distance from the tip of the turnout switch rail to the turnout center position. Based on the turnout switch rail tip mileage information in the ledger, the turnout center mileage is calculated, and the turnout center measurement results are obtained.

[0153] It should be noted that for a detailed description of the railway turnout center measurement system based on airborne LiDAR point cloud provided in the embodiments of the present invention, please refer to the relevant description of the railway turnout center measurement method based on airborne LiDAR point cloud provided in the embodiments of the present invention, which will not be repeated here.

[0154] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A method for measuring the center of a railway turnout based on airborne LiDAR point clouds, characterized in that, The method includes: The target railway turnout and surrounding area were scanned using an airborne LiDAR device to obtain high-density raw 3D point cloud data. The original 3D point cloud data obtained from the scan is preprocessed to obtain high-precision point cloud results. Based on the high-precision point cloud results, according to the spatial geometric characteristics of the railway turnout structure, straight and curved main rail sections in the turnout area are selected, and the centerline coordinate points of the straight and curved main rails are extracted. Based on the coordinates of the centerline of the straight track, the centerline of the straight track is obtained by fitting according to the principle of minimizing track shifting; based on the coordinates of the centerline of the curved track, the curve segment is automatically fitted and optimized with reference to the curve parameters in the ledger, and the centerline of the curved track is obtained by fitting. The virtual intersection of the centerline of the curve basic rail and the centerline of the straight basic rail is obtained to get the coordinates of the turnout center position. Calculate the distance from the tip of the turnout switch rail to the center of the turnout, and estimate the center of the turnout mileage based on the mileage information of the tip of the turnout switch rail in the ledger, thus obtaining the center of the turnout measurement results.

2. The method for measuring the center of a railway turnout based on airborne LiDAR point clouds as described in claim 1, characterized in that, Airborne LiDAR equipment was used to scan the target railway turnout and surrounding area to obtain high-density raw 3D point cloud data, specifically including: Base station control networks and target control networks are pre-deployed along the railway line; the base station control network is used for the construction of UAV flight base stations to meet the requirements of subsequent PPK calculation; the target control network is used for point cloud correction to improve the absolute accuracy of point cloud results. The base station control network prioritizes the use of the existing control network. When the location and accuracy of the existing control network do not meet the requirements for base station construction, a separate base station control network will be deployed. The target control network is set up manually, along both sides of the railway land boundary. The point markers are prefabricated target plates. The three-dimensional coordinates of the target control points are measured using GNSS rapid static and leveling methods. When drones fly, the flight path is laid along both sides of the railway line to avoid entering the railway land boundary, and the flight path overlap and radar parameters are set reasonably.

3. The method for measuring the center of a railway turnout based on airborne LiDAR point clouds as described in claim 1, characterized in that, The raw 3D point cloud data obtained from the scan is preprocessed to obtain high-precision point cloud results, specifically including: The initial trajectory line file is obtained by fusing and solving airborne GNSS data, inertial navigation data, and ground base station GNSS data using integrated navigation solution software. The initial trajectory file is corrected by using target control points deployed along the line to obtain a high-precision trajectory line that is consistent with the coordinates of the target control network. The corrected high-precision trajectory line is fused with the scanned data again to obtain the point cloud results in the WGS84 coordinate system. The point cloud results in the WGS84 coordinate system are transformed into the engineering independent coordinate system to obtain high-precision point cloud results.

4. The method for measuring the center of a railway turnout based on airborne LiDAR point clouds as described in claim 1, characterized in that, Based on high-precision point cloud results, and according to the spatial geometric characteristics of railway turnout structures, straight and curved main rail sections in the turnout area are selected, and the centerline coordinate points of the straight and curved main rails are extracted, specifically including: For straight and curved basic track segments, the methods for extracting the coordinate points of the track centerline include: Based on the high-precision point cloud results, combined with the pre-calibrated structural parameters of the equipment, the laser point cloud on the track surface is automatically classified by the algorithm. Based on the classified track surface laser point cloud, POS position and attitude information and equipment calibration parameters, a local coordinate system of the cross-section track surface center at any mileage is constructed. Under the corresponding local coordinate system, the laser point cloud of the rail top area is filtered according to a preset spatial range to obtain the rail head point data of any cross section. Select equally spaced cross-sectional rail head point cloud data and register them with the preset standard rail head model. Then, determine the left and right rail vertices of each cross section using the registered and fitted rail head model. The average plane coordinates of the left and right rail vertices are taken as the plane coordinates of the centerline point of the track, and the minimum elevation of the left and right rail vertices is taken as the elevation of the centerline point of the track.

5. The method for measuring the center of a railway turnout based on airborne LiDAR point clouds as described in claim 4, characterized in that, The formulas for calculating the plane coordinates and elevation of the centerline points of the track are as follows: in, , , These are the plane X-coordinate, plane Y-coordinate, and elevation of the left rail vertex, respectively. , , These are the plane X-coordinate, plane Y-coordinate, and elevation of the right-side vertex, respectively. , , These are the plane X-coordinate, plane Y-coordinate, and elevation of the centerline point of the track.

6. The method for measuring the center of a railway turnout based on airborne LiDAR point clouds as described in claim 1, characterized in that, Based on the coordinates of the centerline of the basic straight track, and following the principle of minimizing track shifting, the centerline of the basic straight track is fitted, specifically including: The orthogonal least squares fitting algorithm is used for line fitting. The relevant parameters of the line equation are solved based on the criterion of minimizing the sum of squared residuals of the orthogonal distances from the measurement points to the fitted line.

7. The method for measuring the center of a railway turnout based on airborne LiDAR point clouds as described in claim 1, characterized in that, Based on the coordinates of the basic centerline of the curve, the curve segment is automatically fitted and optimized with reference to the parameters of the ledger curve to obtain the basic centerline of the curve, specifically including: When fitting a curve, the curve is first roughly segmented. The basic curve track consists of "straight line-curve-straight line-curve-straight line". The slope of each two adjacent coordinate points of the centerline of the basic curve track is calculated. Based on the change in slope, the boundary between the straight line segment and the curve segment of the centerline of the basic curve track is roughly distinguished. For approximate straight line segments, an orthogonal least squares fitting algorithm is used to fit the straight line; then, circular curves and transition curves are fitted. After the fitting is completed, the coordinates of the straight / transition / transition point and the circular / transition point are calculated; the calculation is repeated multiple times until the coordinate error is lower than the preset threshold, at which point the iteration stops.

8. The method for measuring the center of a railway turnout based on airborne LiDAR point clouds as described in claim 1, characterized in that, The virtual intersection of the centerline of the curved track and the centerline of the straight track is obtained to determine the coordinates of the turnout center. Specifically, this includes: Connect the transition point and the straight-transition point obtained after fitting the centerline of the basic curve track, and then extend it towards the center of the fork, intersecting with the centerline of the basic straight track obtained by fitting. The plane coordinates of the center of the fork are the coordinates of the intersection point. The elevation of the center of the fork is obtained by interpolating the elevations of the two adjacent coordinate points before and after the intersection point on the centerline of the basic straight track.

9. The method for measuring the center of a railway turnout based on airborne LiDAR point clouds as described in claim 1, characterized in that, Calculate the distance from the tip of the turnout switch rail to the turnout center position. Based on the turnout switch rail tip mileage information in the ledger, calculate the turnout center mileage to obtain the turnout center measurement results, specifically including: The position of the turnout switch tip is manually identified from the point cloud and projected onto the fitted basic rail centerline. Then, the distance between the projection point and the turnout center is calculated, and the mileage of the turnout switch tip is checked in the track maintenance log to deduce the mileage information of the turnout center.

10. A railway turnout center measurement system based on airborne LiDAR point clouds, characterized in that, The system includes: The airborne point cloud data acquisition module is used to scan the target railway turnout and surrounding area using airborne LiDAR equipment to acquire high-density raw three-dimensional point cloud data. The airborne point cloud data preprocessing module is used to preprocess the raw 3D point cloud data obtained by scanning to obtain high-precision point cloud results. The turnout basic rail centerline extraction module is used to select straight and curved basic rail segments in the turnout area based on high-precision point cloud results and the spatial geometric characteristics of the railway turnout structure, and extract the centerline coordinate points of the straight and curved basic rails. The turnout area centerline fitting module is used to fit the centerline of the straight track based on the coordinate points of the centerline of the straight track, according to the principle of minimizing the track shifting amount; and to automatically fit the centerline of the curve segment based on the coordinate points of the centerline of the curve, and optimize it with reference to the curve parameters in the ledger, so as to fit the centerline of the curve. The turnout center coordinate calculation module is used to obtain the virtual intersection point of the straight line between the centerline of the curve basic rail and the centerline of the straight basic rail to obtain the coordinates of the turnout center position. The turnout center mileage calculation and output module is used to calculate the distance from the tip of the turnout switch rail to the turnout center position. Based on the turnout switch rail tip mileage information in the ledger, the turnout center mileage is calculated, and the turnout center measurement results are obtained.

Images