Rapid Detection Method for Platform Clearance of Rail Transit Based on 3D Laser Scanning

Abstract

The present invention discloses a rapid detection method for the platform clearance of rail transit based on three-dimensional laser scanning, including: 1) obtaining three-dimensional laser point cloud data of the track, the adjacent station platform and the surrounding buildings by using a self-mobile three-dimensional laser scanning device; 2) identifying and extracting the three-dimensional laser point cloud of the rail; 3) constructing the three-dimensional models of the left and right rails; 4) calculating the three-dimensional coordinates of the line center line; 5) performing plane fitting on the top surface and the inner side surface of the station platform; 6) extracting the outer edge line of the station platform; 7) calculating the horizontal distance and the elevation difference between the outer edge of the station platform and the track center; 8) taking the end point of the outer edge line segment of the platform as the initial vertex of the next segment, and repeating S5-S7 to complete the comprehensive detection of the station platform clearance. This method has high detection efficiency, high precision and reliable detection results.

Description

Technical Field

[0001] The present invention relates to the field of rail transit clearance detection, and in particular to a rail transit platform clearance rapid detection method based on three-dimensional laser scanning. Background Art

[0002] Limits refer to the outline dimension lines that are not allowed to be exceeded for locomotives and vehicles and buildings and equipment facilities close to the line in order to ensure the safety of locomotives and vehicles running on rail transit lines and prevent locomotives and vehicles from colliding with buildings and equipment facilities on adjacent lines. Accurate detection and safety warning of limits are effective means to ensure the safety of train operation, and have important theoretical significance and application value for improving the operational capacity of rail transit. Platform limits refer to the horizontal distance and height difference between the outer eaves of the station platform and the center of the line. Compared with the limits of the line section (roadbed, bridges, tunnels), platform limits must not only ensure the safe distance for locomotives and vehicles to stop, but also improve the comfort of passengers getting on and off the train. Therefore, the station platform limits have more stringent requirements and more precise detection accuracy in the design stage.

[0003] Due to construction technology and management communication problems during the construction period, platform outer wall intrusion accidents occurred frequently. It was found many times that the distance between the side of the platform wall and the center of the line did not meet the limit requirements, and the top of the platform wall was higher than the limit requirements. There was even a safety accident during the joint commissioning and testing when the high platform cap of the station scratched two test vehicles. The occurrence of tragic accidents is not accidental. The reason is still due to the lack of efficient and comprehensive technical means for platform limit detection. The traditional platform limit measurement method is mainly contact measurement, such as the cross-section method, the comprehensive section method and the trajectory method. They are all contact detection, which are characterized by requiring a lot of manual operation, and there are significant problems such as large workload, low efficiency, low reliability and poor accuracy. Limit detection based on cross-section camera method belongs to non-contact measurement. This method improves the degree of automation of limit detection, but it is greatly disturbed by light and has high requirements for the working environment. With the development of laser measurement technology, three-dimensional laser scanning technology, as a new non-contact measurement method, has the advantages of fast, high precision and all-weather measurement, and is suitable for operating line measurement during skylight time. However, 3D laser scanning technology can only quickly obtain the entire 3D point cloud of the station platform. It is still necessary to combine the station platform characteristics and limit detection content, study the laser point cloud data processing algorithm, and develop corresponding data processing software to realize the rapid detection of rail transit station platform limits based on 3D scanning technology to meet the rapidly developing rail transit safe transportation needs. Summary of the invention

[0004] In order to solve the problems existing in the prior art, the present invention provides a rail transit platform limit rapid detection method based on three-dimensional laser scanning, which can quickly and accurately detect station platform limits.

[0005] To this end, the present invention adopts the following technical solutions:

[0006] A rapid detection method for the platform clearance of rail transit based on three-dimensional laser scanning, comprising the following steps:

[0007] S1, Acquisition of the three-dimensional laser point cloud of the station: Use self-mobile three-dimensional laser scanning equipment to acquire the three-dimensional laser point cloud data of the track, the adjacent station platform and the surrounding buildings;

[0008] S2, Identification and extraction of the three-dimensional laser point cloud of the rail: Use the relatively stable geometric position relationship of the rail point cloud coordinates in the coordinate system of the self-mobile three-dimensional laser scanning equipment to automatically identify and extract the rail point cloud;

[0009] S3, Construction of the three-dimensional models of the left and right rails: Use the rail point cloud identified and extracted in step S2, combined with the geometric dimensions of the rail model, and adopt the point cloud fitting method to perform three-dimensional reconstruction of the rail model;

[0010] S4, Calculation of the three-dimensional coordinates of the line center line: Use the three-dimensional models of the left and right rails obtained in S3, and based on the prior knowledge of the line design parameters, automatically calculate the three-dimensional coordinates of the line track center line;

[0011] S5, Plane fitting of the top surface and the inner side surface of the station platform: Based on the laser point cloud data obtained by scanning the platform, automatically segment the laser point cloud of the top surface and the inner side surface of the station platform, and fit the inner side and the top plane of the outer eaves of the station platform;

[0012] S6, Extraction of the edge line of the outer eaves of the station platform: According to the inner side and the top plane of the outer eaves of the station platform fitted in S5, use the plane intersection algorithm in formula (5) to realize the extraction calculation of the edge line of the outer eaves of the platform. The geometric expression of the spatial plane is (P, N), where P is the three-dimensional coordinate of a certain point on the plane, and N is the normal vector of the plane;

[0013]

[0014] In the formula:

[0015] (P1, N1): is the geometric expression of the inner side plane of the station platform;

[0016] (P2, N2): is the geometric expression of the top plane of the station platform;

[0017] N3: is the normal vector of the edge line of the outer eaves of the platform;

[0018] X: is the vertex on the edge line of the outer eaves of the platform;

[0019] S7. Calculation of the horizontal distance and elevation difference between the outer eaves of the station platform and the track center: According to the three-dimensional coordinates of the vertices of the center line of the track and the outer edge line of the station platform, use formula (1) to calculate the horizontal distance (L i ) and elevation difference (H i ) between the outer eaves of the station platform and the track center line;

[0020] S8. Take the end point of the outer eaves edge line obtained in step S6 as the initial vertex of the next segment, repeat steps S5 - S7 until the full inspection of the platform clearance within the station is completed, and determine whether there is an encroachment situation according to the allowable value of the designed platform clearance, and form an inspection report.

[0021] Among them, step S5 includes the following sub-steps:

[0022] S51. Obtaining the initial values of the horizontal distance and elevation difference between the outer eaves of the station platform and the track center: Manually select a point P T from the laser points of the outer eaves of the platform in the station point cloud, and then select the center point P L of the top of the left rail and the center point P R of the top of the right rail respectively from the corresponding cross-section, and use formula (1) to calculate the initial value of the distance from the track center to the outer eaves of the platform:

[0023]

[0024] L0 = Vector2D(P T - (P L + P R ) / 2.0) (1)

[0025] In the formula:

[0026] L0: The initial value of the horizontal distance between the outer eaves of the station platform and the track center;

[0027] H0: The initial value of the elevation difference between the outer eaves of the station platform and the track center;

[0028] The elevation value of the laser point of the outer eaves of the platform;

[0029] The elevation value of the laser point at the top of the left rail;

[0030] The elevation value of the laser point at the top of the right rail;

[0031] Vector2D(P T - (P L + P R ) / 2.0): The horizontal distance between the laser point of the outer eaves of the platform and the track center;

[0032] S52, Laser point cloud segmentation of the top surface and inner side of the station platform: The width of the top surface of the station platform is set as ΔS, the height of the side surface of the station platform is set as ΔV, and the distance of the station platform along the mileage direction is set as ΔD. Using the platform outer eaves vertex P selected through manual interaction in step S51 T as the cube vertex, with ΔS, ΔV, and ΔD as the side lengths and δ as the threshold of the side length, construct a cube bounding box;

[0033] S53, Determine whether the laser point cloud is within the cube bounding box: By traversing the station laser point cloud, judge whether it is within the cube, so as to segment the laser point cloud of the platform outer eaves from all the station laser point clouds;

[0034] S54, Plane fitting calculation of the top surface and inner side of the station platform: Use the plane fitting algorithm to fit the planes of the top and inner sides of the platform outer eaves respectively, obtain the normal vector and constant of the fitted plane, and calculate the residual value of the laser point cloud plane fitting.

[0035] In step S52, construct the cube bounding box according to formula (2):

[0036]

[0037] In the formula:

[0038] BBox: Cube bounding box;

[0039] minx, miny, minz: The minimum values of the vertex coordinates of the cube bounding box;

[0040] maxx, maxy, maxz: The maximum values of the vertex coordinates of the cube bounding box;

[0041] P Tx 、P Ty 、P Tz : Coordinate values of the laser point cloud of the platform outer eaves selected through manual interaction.

[0042] The formula for judging whether the laser point cloud is within the cube in step S53 is as follows:

[0043]

[0044] In the formula:

[0045] Pi: The i-th laser point;

[0046] Pi x 、Pi y 、Pi z : The x, y, and z coordinate values of the i-th laser point;

[0047] Segment Plateform, Segment Other : They are the point cloud of the platform outer eaves and the other point clouds except the point cloud of the platform outer eaves respectively;

[0048] BBox minx , BBox miny , BBox minz : They are the minimum values of the vertex coordinates of the cubic bounding box respectively;

[0049] BBox maxx , BBox maxy , BBox maxz : They are the maximum values of the vertex coordinates of the cubic bounding box respectively;

[0050] δ: The filtering threshold of the cubic bounding box, with the unit of meter.

[0051] In step S54, plane fitting is performed using formula (4):

[0052] N x *x + N y *y + N z *z + constVal = 0 (4)

[0053] In the formula:

[0054] N x , N y , N z : The normal vector of the fitting plane;

[0055] constVal: The constant value of the fitting plane.

[0056] The present invention has the following beneficial effects:

[0057] 1. The detection method of the present invention realizes the rapid segmentation of a small amount of effective laser point cloud data of the platform outer eaves from the massive point clouds scanned by the station by using the cubic bounding box segmentation algorithm, significantly improving the data processing efficiency. Using the method of the present invention to detect the platform limit of a 450m long platform, the data automatic processing time only needs 2 minutes;

[0058] 2. The present invention adopts the plane fitting algorithm and the plane intersecting line algorithm to realize the high-precision calculation of the horizontal distance and elevation difference between the platform outer eaves and the center line of the line. The plane fitting algorithm can effectively eliminate the influence of noise point clouds and improve the accuracy of the detection algorithm;

[0059] 3. The present invention uses the scanned laser point cloud to realize the three-dimensional reconstruction of the rail and the platform outer eaves through the point cloud fitting method, and then extracts the center line of the rail and the outer eaves line of the platform from the reconstructed model. This method does not directly measure the coordinates of the platform outer eaves from the center of the scanner, avoiding the influence of scanner jitter on the detection results and ensuring the reliability of the detection results;

[0060] 4. Compared with the existing point-by-point measurement method, the detection method of the present invention is faster, more accurate and more comprehensive, which helps to expand the application of three-dimensional laser scanning technology in the field of rail transit clearance detection and improve the rapid detection level of rail transit. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is a flowchart of the detection method of the present invention;

[0062] Figure 2 is the laser point cloud data within the scope of a rail transit station;

[0063] Figure 3 is a self-mobile laser scanning system and its coordinate system;

[0064] Figure 4 is a three-dimensional model of a rail constructed according to the standard cross-section diagram of the rail;

[0065] Figure 5 is a schematic diagram of calculating the center line of the line based on the left and right rail models;

[0066] Figure 6 is a detailed dimension diagram of the construction clearance of a passenger dedicated line railway. DETAILED DESCRIPTION OF THE INVENTION

[0067] The following combines the drawings and specific embodiments to elaborate in detail on the rapid detection method for the platform clearance of rail transit based on three-dimensional laser scanning of the present invention.

[0068] Embodiment

[0069] As Figure 1 shown, the rapid detection method for the platform clearance of rail transit based on three-dimensional laser scanning of the present invention includes the following steps:

[0070] S1. Obtain the three-dimensional laser point cloud of the station: Use a self-mobile three-dimensional laser scanning system to quickly obtain the three-dimensional laser point cloud data of the track, the adjacent station platform and the surrounding buildings. The obtained laser point cloud data of the rail transit station is as Figure 2 shown. The type of rail used for this rail transit line is 60 kg / m rail (60 rail), and the length of the detected platform is about 450 meters.

[0071] See Figure 3, the self - mobile 3D laser scanning system integrates modules such as an odometer, a high - precision laser scanner, an inclinometer, a gauge measurement system, a high - precision time synchronization module, and a programmable logic controller (PLC) on a self - mobile rail car platform. It can quickly complete 3D scanning measurements in the station, featuring high efficiency, high precision, miniaturization, automation, and multi - functionality. The self - mobile 3D laser scanning detection system automatically travels on the rail transit line, with a speed setting range of 0 - 5 km / h, a measurement range in the direction perpendicular to the track of 0.3 - 200 m, a measurement accuracy better than 5 mm, and 1 million laser points are collected per second.

[0072] The laser scanner integrated in the self - mobile laser scanning system uses a 2D cross - section and 360° circumferential scanning method, and the coordinate origin of the scanning cross - section is at the center position of the scanner. Set the forward driving direction of the scanner (i.e., the line mileage direction) as the X - axis, the vertical ground direction as the Z - axis, and the Y - axis is perpendicular to the X - axis and the Z - axis respectively (i.e., horizontally perpendicular to the line direction). A right - hand coordinate system is established with the XYZ axes. The mileage direction is the X - coordinate value of the point cloud, the horizontal distance of the scanner is the Y - coordinate value, and the vertical distance is the Z - coordinate value, obtaining the 3D laser point cloud data of the station area.

[0073] S2, Identification and extraction of 3D laser point cloud of rail: The automatic identification and extraction of the rail point cloud are carried out by using the relatively stable position relationship of the rail point cloud coordinates in the coordinate system of the self - mobile laser scanning system. In this embodiment, height filtering and rectangular segmentation methods are used to automatically extract the laser point cloud of the rail.

[0074] S3, Construct 3D models of the left and right rails according to the standard cross - section diagram of 60 kg / m rails. The length of the model is the same as the length of the track line segment, which is 1 m, as Figure 4 shown. Fit and register the standard rail model with the rail point cloud obtained in S2 to reconstruct the 3D models of the left and right rails of the line.

[0075] S4, Using the 3D models of the left and right rails obtained in S3, based on the line design parameters, calculate the 3D coordinates of the center line of the track of the line to obtain information such as the distance of the actual track and the direction of the track; according to the definition of the reference track, after determining the reference track, then based on the reference track, horizontally offset 1 / 2 of the standard gauge (the standard gauge defined in China is 1435 mm) to the other rail, calculate the center line points of the track in segments, and then connect these segmented center line points of the track in sequence to form the line center line. Finally, smooth the formed line center line to obtain a relatively smooth line center line, as Figure 5 shown.

[0076] S5, Plane Fitting of the Top Surface and Inner Side Surface of the Station Platform: Based on the laser point cloud data obtained by scanning the platform, fit the planes of the top surface and inner side surface of the station platform. The coordinates of the platform eaves obtained in this step are segmented extraction values, and the segment length can be set according to actual needs. The segment length in this embodiment is set to 1.0 m (i.e., each fitting is 1.0 m along the line direction). The specific steps are as follows:

[0077] S51, Obtaining the Initial Values of the Horizontal Distance and Elevation Difference between the Station Platform Eaves and the Track Center: There are fixed design values for the spatial positions of the station platform eaves and the track center. Due to construction errors or position adjustments of the track during later operation, there are small-scale continuous changes in the position relationship. The initial values can be calculated using the following method: Manually select a point (P T ) from the laser points of the platform eaves in the station point cloud, and then select the center point of the top of the left rail (left rail P L ) and the center point of the top of the right rail (right rail P R ) from the corresponding cross-section respectively. Use formula (1) to calculate the initial value of the distance from the track center to the platform eaves.

[0078]

[0079] L0 = Vector2D(P T - (P L + P R ) / 2.0) (1)

[0080] In the formula:

[0081] L0: The initial value of the horizontal distance between the station platform eaves and the track center;

[0082] H0: The initial value of the elevation difference between the station platform eaves and the track center;

[0083] The elevation value of the laser point of the platform eaves;

[0084] The elevation value of the laser point at the top of the left rail;

[0085] The elevation value of the laser point at the top of the right rail;

[0086] Vector2D(P T - (P L + P R ) / 2.0): The horizontal distance between the laser point of the platform eaves and the centers of the left and right rails.

[0087] The initial values of the horizontal distance and elevation difference between the station platform eaves and the track center are generally selected at the starting position of the station scan.

[0088] S52, Laser point cloud segmentation of the top surface and inner side surface of the station platform: The width of the top surface of the station platform is set as ΔS, the height of the side surface of the station platform is set as ΔV, and the distance of the station platform along the mileage direction is set as ΔD. Taking the platform outer eaves vertex P selected by manual interaction in step S51 T as the cube vertex, with ΔS, ΔV, and ΔD as the side lengths, and δ as the threshold of the side length, construct a cube bounding box according to formula (2):

[0089]

[0090] In the formula:

[0091] BBox: Cube bounding box;

[0092] minx, miny, minz: The minimum values of the vertex coordinates of the cube bounding box;

[0093] maxx, maxy, maxz: The maximum values of the vertex coordinates of the cube bounding box;

[0094] P Tx 、P Ty 、P Tz : The coordinate values of the laser points of the platform outer eaves selected by manual interaction.

[0095] In this embodiment, the width ΔS of the platform top surface is set to 0.5 m, ΔV takes the designed height value of 0.3 m of the inner side surface of the platform, and the distance ΔD in the mileage direction is set to 0.5 m, which is equal to the segment length in step S5.

[0096] S53, Determine whether the laser point cloud is inside the cube bounding box: By traversing the station laser point cloud, judge whether it is inside the cube, so as to segment the laser point cloud of the platform outer eaves from all the station laser point clouds. The method for judging whether the laser point cloud is inside the cube is shown in formula (3).

[0097]

[0098] In the formula:

[0099] Pi: The i-th laser point;

[0100] Pi x 、Pi y 、Pi z : The x, y, and z coordinate values of the i-th laser point;

[0101] Segment Plateform 、Segment Other : Respectively the platform outer eaves point cloud and other point clouds;

[0102] BBox minx, BBox miny , BBox minz : They are respectively the minimum values of the vertex coordinates of the cubic bounding box;

[0103] BBox maxx , BBox maxy , BBox maxz : They are respectively the maximum values of the vertex coordinates of the cubic bounding box;

[0104] δ: The filtering threshold of the cubic bounding box, with the unit of meter.

[0105] In order to segment the laser point cloud outside all platform eaves, the threshold δ in this embodiment is set to 0.03m.

[0106] S54, Plane fitting calculation of the top surface and inner side of the station platform: Considering the construction error and deformation during operation, the top surface and inner side of the platform cannot be strictly planes. Use the plane fitting algorithm to fit the planes at the top and inner sides of the platform eaves respectively, obtain the normal vector and constant of the fitted plane, and calculate the distance residuals of all laser point clouds to the fitted plane to judge the plane fitting accuracy and effect. The plane fitting formula is as follows:

[0107] N x *x + N y *y + N z *z + constVal = 0 (Equation (4)) where:

[0108] N x , N y , N z : The normal vector of the fitted plane;

[0109] constVal: The constant value of the fitted plane;

[0110] x, y, z: The three-dimensional coordinates of the laser point.

[0111] S6, Extraction of the edge line of the station platform eaves: According to the planes fitted to the inner side and top of the station platform eaves in step S54, use the plane intersection algorithm in formula (5) to implement the calculation of the extraction of the edge line of the platform eaves. Formula (4) in step S54 is the general algebraic expression of the spatial plane. Another geometric expression of the spatial plane is (P, N), where P is the three-dimensional coordinate of a point on the plane and N is the normal vector of the plane;

[0112]

[0113] Where:

[0114] (P1, N1): The geometric expression of the inner plane of the station platform;

[0115] (P2, N2): It is the geometric expression of the top plane of the station platform;

[0116] N3: It is the normal vector of the outer edge line of the platform;

[0117] X: It is the vertex on the outer edge line of the platform.

[0118] In this embodiment, the length of the outer edge line of the platform is the same as the length of ΔD in step S52, which is 0.5 m.

[0119] S7. Calculation of the horizontal distance and elevation difference between the outer edge of the station platform and the center of the track: According to the three-dimensional coordinates of the vertices of the center line of the line track and the outer edge line of the station platform, use formula (1) to calculate the horizontal distance (L i ) and elevation difference (H i ) of the station platform edge;

[0120] In this embodiment, project the midpoint P of the outer edge line segment of the platform to the point P' on the center line of the track. Use P and P', and according to formula (1), calculate the horizontal distance (L i ) and the elevation difference (H i ) between the outer edge of the station platform and the center of the track.

[0121] S8. Take the end point of the outer edge line of the platform obtained in step S6 as the initial vertex of the next segment. Repeat steps S5 - S7 until the full detection of the platform clearance within the station is completed. And according to the allowable value of the designed platform clearance, as Figure 6 shown, determine whether there is an encroachment situation and form a detection report.

[0122] The historical data of the clearance detection can be established in a database, and the horizontal distance (L i ) and the elevation difference (H i ) change curves at the same mileage can be drawn to find the change trend of the platform clearance. When the change trend is significant, engineering rectification measures should be taken in advance.

[0123] Using the method of the present invention, it only takes 2 minutes to process the laser point cloud data of a 450 - m - long rail transit platform, greatly improving the detection efficiency.

[0124] Experimental accuracy analysis:

[0125] Two methods are adopted to analyze the accuracy of the boundary measurement results of the present invention: First, the results of round-trip scanning detection are compared by using the method of the present invention, and the differences between round-trip measurements are all less than 3 mm; Second, the "DJJ-8 boundary laser detector" developed by Jinan Landong Laser Technology Co., Ltd., Laser Research Institute of Shandong Academy of Sciences is used to verify the accuracy of the method of the present invention. The "DJJ-8 boundary laser detector" has been widely used and recognized in domestic boundary detection, and the measurement accuracy of this instrument is ±3 mm. The measurement results of the detection method of the present invention are compared with the results detected by the above DJJ-8 boundary laser detector, and the differences are all less than 5 mm. The comparison results are statistically shown in the following table:

[0126]

[0127] From the above results, it can be seen that the measurement accuracy of the present invention meets the specification requirements of the current "Measurement and Data Format of Railway Building Actual Boundary" (TBT 3308-2013).

Claims

1. A rapid detection method for the platform clearance of rail transit based on 3D laser scanning, comprising the following steps: S1, Acquisition of 3D laser point cloud of the station: Use self-mobile 3D laser scanning equipment to acquire 3D laser point cloud data of the track, adjacent station platform and surrounding buildings; S2, Identification and extraction of 3D laser point cloud of the rail: Utilize the relatively stable geometric position relationship of the rail point cloud coordinates in the coordinate system of the self-mobile 3D laser scanning equipment to automatically identify and extract the rail point cloud; S3, Construction of 3D models of the left and right rails: Use the rail point cloud identified and extracted in step S2, combined with the geometric dimensions of the rail model, and adopt the point cloud fitting method to perform 3D reconstruction of the rail model; S4, Calculation of 3D coordinates of the line centerline: Use the 3D models of the left and right rails obtained in S3, and based on the prior knowledge of the line design parameters, automatically calculate the 3D coordinates of the line track centerline; S5, Plane fitting of the top surface and inner side surface of the station platform: Based on the laser point cloud data obtained by scanning the platform, automatically segment the laser point cloud of the top surface and inner side surface of the station platform, and fit the inner side and top planes of the outer eaves of the station platform; S6, Extraction of the edge line of the outer eaves of the station platform: According to the inner side and top planes of the outer eaves of the station platform fitted in S5, use the plane intersection algorithm in formula (5) to realize the calculation of the extraction of the edge line of the outer eaves of the platform. The geometric expression of the space plane is (P, N), where P is the 3D coordinate of a certain point on the plane and N is the normal vector of the plane; In the formula: (P1, N1): is the geometric expression of the inner side plane of the station platform; (P2, N2): is the geometric expression of the top plane of the station platform; N3: is the normal vector of the edge line of the outer eaves of the platform; X: is the vertex on the edge line of the outer eaves of the platform; Calculation of the horizontal distance and elevation difference between the outer eaves of the station platform and the track center: Based on the three-dimensional coordinates of the vertices of the center line of the line track and the outer edge line of the station platform, calculate the horizontal distance L between the outer eaves of the station platform and the track center line i and the elevation difference H i ; S8, Take the end point of the edge line of the outer eaves of the platform obtained in step S6 as the initial vertex of the next segment, repeat steps S5 - S7 until the comprehensive detection of the platform clearance within the station is completed, and determine whether there is an encroachment situation according to the allowable value of the designed platform clearance, and form a detection report.

2. The rapid detection method for the platform clearance of rail transit based on three-dimensional laser scanning according to claim 1, wherein, Step S5 Comprises the following sub-steps: S51, Obtaining the initial value of the horizontal distance and elevation difference between the outer eaves of the station platform and the track center: Manually select a point P from the laser points of the outer eaves of the platform in the station point cloud T , and then select the center point P of the top of the left rail L and the center point P of the top of the right rail R respectively from the corresponding cross-section, and use formula (1) to calculate the initial value of the distance from the track center to the outer eaves of the platform: L0 = Vector2D(P T -(P L +P R ) / 2.0) (1) In the formula: L0: Initial value of the horizontal distance between the outer eaves of the station platform and the track center; H0: Initial value of the elevation difference between the outer eaves of the station platform and the track center; Elevation value of the laser points on the outer eaves of the platform The elevation value of the laser point at the top of the left rail; Elevation value of the laser point on the top of the right rail; Vector2D(P T -(P L +P R ) / 2.0): The horizontal distance between the laser point on the outer eaves of the platform and the track center; S52, Laser point cloud segmentation of the top and inner side of the station platform: The width of the top surface of the station platform is set as ΔS, the height of the side surface of the station platform is set as ΔV, and the distance of the station platform along the mileage direction is set as ΔD. Using the platform outer edge vertex P selected by manual interaction in step S51 T as the cube vertex, with ΔS, ΔV, and ΔD as the side lengths and δ as the threshold of the side length, construct a cube bounding box; S53, Determine whether the laser point cloud is within the cubic bounding box: By traversing the laser point cloud of the station, judge whether it is within the cube, so as to segment the laser point cloud of the outer eaves of the platform from all the laser point clouds of the station; S54, Plane fitting calculation of the top surface and inner side surface of the station platform: Use the plane fitting algorithm to fit the planes of the top and inner sides of the outer eaves of the platform respectively, obtain the normal vector and constant of the fitted plane, and calculate the residual value of the plane fitting of the laser point cloud.

3. The rapid detection method for the track traffic platform clearance based on 3D laser scanning according to claim 2, wherein In step S52, construct the cubic bounding box according to formula (2): In the formula: BBox: Cubic bounding box; minx, miny, minz: are the minimum values of the vertex coordinates of the cubic bounding box; maxx, maxy, maxz: are the maximum values of the vertex coordinates of the cubic bounding box; P Tx 、P Ty 、P Tz : Manually interactively select the coordinate values of the laser point cloud on the outer eaves of the platform.

4. The rapid detection method for the track traffic platform clearance based on three-dimensional laser scanning according to claim 2, characterized in that The formula for judging whether the laser point cloud is within the cube in step S53 is as follows: In the formula: Pi: The i-th laser point; Pi x 、 Pi y 、 Pi z : The x, y, and z coordinate values of the i-th laser point; Segment Plateform 、Segment Other : They are the point cloud of the outer eaves of the platform and the other point clouds except the point cloud of the outer eaves of the platform respectively; BBox minx and BBox miny and BBox minz : They are respectively the minimum values of the vertex coordinates of the cubic bounding box; BBox maxx and BBox maxy and BBox maxz : They are the maximum values of the vertex coordinates of the cubic bounding box, respectively. δ: The filtering threshold of the cubic bounding box, with the unit of meter.

5. The rapid detection method for the track traffic platform clearance based on 3D laser scanning according to claim 2, wherein In step S54, plane fitting is performed using formula (4): N x *x + N y *y + N z *z + constVal = 0 (Equation (4)) Where: N x 、N y 、N z : The normal vector of the fitting plane; constVal: The constant value of the fitted plane.

Images