Railway vehicle running posture detection method and detection system

Abstract

The present application belongs to the field of track vehicle posture detection, and particularly relates to a track vehicle running posture detection method and a detection system. The track vehicle running posture detection method and the detection system utilize the characteristics that the distance between linear laser spots on the track of two parallel laser beams is unchanged and the properties of similar triangles, convert the calibration process into only needing to measure the distance between the two laser spots and input the system to obtain the detection result, achieve the effect of calibration-free, solve the problem that the existing detection method cannot be applied on a large scale due to the difficulty in calibration, and realize real-time monitoring of the whole process running posture of the track vehicle.

Description

Technical Field

[0001] This invention belongs to the field of rail vehicle attitude detection, and particularly relates to a method and system for detecting the running attitude of rail vehicles. Background Technology

[0002] With the continuous development of my country's rail vehicle industry, especially in recent years with the increasing intelligence and speed of rail transportation, the safety, stability, and comfort of rail transport have received increasing attention. For freight trains, especially oversized freight trains, the running posture of the car body is related to whether the goods can safely reach their destination and whether it will affect the safety of other trains on the line; while for passenger trains, the running posture of the car body not only affects passenger comfort but is also closely related to passenger safety. Therefore, rail vehicle running posture detection devices are particularly important.

[0003] Excessive attitude angles and lateral and vertical offsets pose significant safety hazards to rail vehicles, especially in adverse weather conditions such as strong winds. Therefore, real-time monitoring of attitude parameters throughout train operation is crucial. Currently, rail vehicle attitude detection methods, both domestically and internationally, can be broadly categorized into two types: one involves installing detection devices on the ground in hazardous sections such as tunnels and crosswind zones; the other involves installing detection devices on the vehicle body for onboard detection. However, most onboard visual inspection devices require calibrating each measurement point using multiple calibration blocks of known thickness. Calibration involves continuously stacking calibration blocks, is prone to errors, and is time-consuming, hindering large-scale application. Therefore, there is an urgent need for an onboard rail vehicle attitude detection method and system that can achieve real-time monitoring of the rail vehicle's attitude throughout its entire operation without calibration. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for detecting the running attitude of rail vehicles, which can be calibration-free and solves the problem that existing detection methods cannot be applied on a large scale due to the difficulty of calibration. It can realize real-time monitoring of the running attitude of rail vehicles throughout the entire process.

[0005] The technical solution of this invention is detailed below:

[0006] Firstly, a method for detecting the running attitude of a rail vehicle includes the following steps:

[0007] S1. Three detection groups, namely the first, second, and third, are set at the bottom of the same carriage of the train above the track. Each detection group includes a laser emission point and a camera mounting point. The laser emission point can continuously emit two parallel laser beams towards the track. The two laser beams illuminate the track surface to form two linear laser spots perpendicular to the center line of the track surface. The camera mounting point is equipped with an area array camera that can clearly capture the linear laser spots. The area array camera uses a fixed-focus lens. The image plane of the area array camera is parallel to the track plane, and the focal points of the area array cameras of the three detection groups are not on a straight line.

[0008] S2. When the train is stationary on the track, it is considered to be in the positive position 0 point, that is, there is no offset on the track. At this time, the distance between the two linear laser spots of each set of measuring points is measured, and the position of the focal point of the area array camera at this time is regarded as the reference position.

[0009] S3. At time t during the train's operation, based on the principles of optical imaging and the properties of similar triangles, calculate the vertical dynamic offset ΔZ and lateral dynamic offset ΔY of the focal point of the area array camera relative to the reference position for each set of measuring points.

[0010] S4. Calculate the five attitude parameters of the train at time t: roll angle α, pitch angle β, yaw angle γ, yaw offset Δy, and buoyancy offset Δz.

[0011] S5. The real-time weight risk coefficient under the current operating environment is S. t The risk factor S is predetermined for the line under the current operating environment. S is calculated based on five attitude parameters. t Comparison S t And S, when S t When S ≥ S, the train's operating posture under this operating environment at time t is unsafe, and a warning is issued; when S ≥ S t If the value is less than S, then the train's operating posture at time t is safe under this operating environment, and no warning is issued.

[0012] Beneficial effects:

[0013] Existing vehicle-mounted vision-based vehicle attitude detection solutions require multiple calibration blocks of known thickness to be continuously stacked and their thicknesses varied after the detection device is installed. This calibration process is prone to errors, and recalibration is necessary when wheelsets or bogies are replaced, wasting significant time and manpower. This invention utilizes the characteristic that the distance between the linear laser spots of two parallel laser beams on the track remains constant and the properties of similar triangles to transform the calibration process into one where only the distance between the two laser spots needs to be measured and input into the system to obtain the detection result, achieving a calibration-free effect.

[0014] Furthermore, in step S1, the plane where the camera mounting points of the three detection groups are located is parallel to the track plane. When applying this invention, it is necessary to measure the lateral distance W and longitudinal distance L between the focal points of the area array cameras. To simplify the measurement process, the lateral and longitudinal distances between the two camera mounting points can be approximated, namely, the lengths of the line segments projected onto the lateral symmetry plane of the vehicle body and the lengths of the line segments projected onto the longitudinal symmetry plane of the vehicle body. By measuring the length of the line connecting the two camera mounting points and the angular deviations relative to the lateral and longitudinal symmetry planes of the vehicle body, the lateral distance W and longitudinal distance L between the focal points of the area array cameras can be calculated. However, the process of measuring the angle is prone to errors. To make the measurement simpler and more reliable, this invention employs an optimized method. See also... Figure 1 When the plane where the camera mounting points of the three detection groups are located is parallel to the track plane, the lateral distance between the camera mounting point 100 of the first detection group and the camera mounting point 200 of the second detection group, and the longitudinal distance between the camera mounting point 100 of the first detection group and the camera mounting point 300 of the third detection group, can be easily obtained by drawing a perpendicular line to the line parallel to the center line of the track where the camera mounting point 100 of the first detection group is located.

[0015] Furthermore, in step S1, the camera mounting points of the first and second detection groups are located at the same end of the carriage, and their straight line is perpendicular to the center line of the track; the camera mounting points of the first and third detection groups are located on the same side of the carriage, and their straight line is parallel to the center line of the track. See also... Figure 2 By further defining the positional relationship of the camera mounting points, the lateral distance between the camera mounting point 100 of the first detection group and the camera mounting point 200 of the second detection group is equal to the line connecting them, and the longitudinal distance between the camera mounting point 100 of the first detection group and the camera mounting point 300 of the third detection group is also equal to the line connecting them, making the measurement simpler and more reliable.

[0016] Furthermore, in step S3, the step of calculating the vertical dynamic offset ΔZ includes:

[0017] Establish a track-following coordinate system that moves forward with the train. The origin O is the projection of the geometric center point of the car body onto the track plane. The positive direction of the X-axis is the center line of the track plane along the direction of train travel. The positive direction of the Z-axis is the direction perpendicular to the track plane and upward. The Y-axis is perpendicular to the OXZ plane and points to the right.

[0018] Calculate the vertical dynamic offset ΔZ1 of the focal point of the first detection group's area array camera at time t, where a and b are two linear laser spots in the first detection group; A and B are the projections of these two linear laser spots onto the OXZ plane, and the distance between them is denoted by L. AB It means, L ABThe distance is equal to the distance between parallel linear laser spots a and b; points C and D are the corresponding imaging points of points A and B on the camera's image plane, and the distance between them is denoted by L. DC Indicates: F is the focal point of the area scan camera; MF and FH are the perpendicular segments from point F to lines AB and CD, respectively; L MF It is the camera's focal length f, L FH It is the distance from the focal point of the area array camera to the track plane. According to the principles of optical imaging and the properties of similar triangles, where:

[0019]

[0020] When the train is at the positive position 0, the apparent vertical offset is 0. It is the distance from the focal point of the area array camera to the track plane at this moment. The distance between points C and D at this moment is:

[0021]

[0022] At time t during the train's operation, It is the distance from the focal point of the area array camera to the track plane at this moment. The distance between points C and D at this moment is:

[0023]

[0024] The vertical dynamic offset ΔZ1 of the focal point of the first detection group's area array camera is:

[0025]

[0026] Similarly, the vertical dynamic offsets ΔZ2 and ΔZ3 of the focal points of the area array cameras in the second and third detection groups were calculated.

[0027] Furthermore, in step S3, the step of calculating the lateral dynamic offset ΔY includes:

[0028] Calculate the lateral dynamic offset ΔY1 of the focal point of the first group of measurement point area array cameras at time t, where point A′ is the midpoint of the track plane and point C′ is the imaging point of point A′ on the image plane, using L C′C The distance between point C and point C′ is represented by L. A′A Let A' represent the distance between point A and point A, where:

[0029]

[0030] When the train is at the positive position 0, the lateral dynamic offset is considered to be 0, and this is used as the measurement standard for the lateral dynamic offset at time t. This is the distance between point A′ and point A at this moment. For the distance between point C and point C′ at this moment, we have:

[0031]

[0032] At time t during the train's operation, This is the distance between point A′ and point A at this moment. For the distance between point C and point C′ at this moment, we have:

[0033]

[0034] The lateral dynamic offset ΔY1 of the focal point of the first detection group's area array camera is:

[0035]

[0036] Similarly, the lateral dynamic offsets ΔY2 and ΔY3 of the focal points of the second and third detection group area array cameras were calculated.

[0037] Furthermore, in step S4, the steps of calculating the train's roll angle α, pitch angle β, and yaw angle γ at time t include:

[0038] W represents the lateral distance between the camera mounting points of the two detection groups on the same head, i.e., the distance between the area array cameras along the Y direction, which is directly measured using a millimeter ruler. The roll angle α is:

[0039]

[0040] L is the longitudinal distance between the camera mounting points of the two detection groups on the same side, that is, the distance between the area array cameras along the X direction, which is directly measured using a millimeter ruler. The dot angle β is:

[0041]

[0042] The head-shaking angle γ is:

[0043]

[0044] Furthermore, in step S4, the formulas for calculating the train's yaw offset Δy and buoyancy offset Δz at time t are as follows:

[0045]

[0046]

[0047] When i = 1, the sign before α is "+", the sign before β is "+", and the sign before γ is "+".

[0048] When i = 2, the sign before α is "-", the sign before β is "+", and the sign before γ is "+".

[0049] When i = 3, the sign before α is "+", the sign before β is "-", and the sign before γ is "-".

[0050] Furthermore, in step S5, S t The calculation formula is as follows:

[0051] S t =p1|α|+p2|β|+p3|γ|+p4|Δy|+p5|Δz|

[0052] Among them, different weights are set for each attitude parameter for different operating environments, {p i , i∈(1,5)} are the weight parameters of five different attitude factors.

[0053] Secondly, a rail vehicle operating attitude detection system includes:

[0054] The detection device comprises at least three sets, each set including a laser emitter, a prism box, and an area array camera. The laser emitter is capable of continuously emitting two parallel laser beams toward the track, which illuminate the track to form a linear laser spot. The area array camera is capable of continuously capturing the linear laser spot of the same set of detection devices and sending the data to the calculation and judgment device in real time after obtaining a clear image.

[0055] The calculation and judgment device acquires data from the area array camera of the detection device, calculates the roll angle α, pitch angle β, yaw angle γ, yaw offset Δy, and buoyancy offset Δz of the train at time t, and determines whether the train's running posture is safe, and sends a signal to the signal output device.

[0056] A signal output device is used to receive instructions from the calculation and judgment device and issue a warning signal.

[0057] Furthermore, the detection device includes a detection box, which comprises an area array camera, a laser emitter, and a prism box. The laser emitted by the laser emitter passes through the prism box, transforming from a single laser beam into two parallel laser beams. This combination of a laser emitter and a prism box ensures that the two laser beams are parallel to each other.

[0058] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0059] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0060] Figure 1 This is a schematic diagram of the optimized camera mounting point distribution for the rail vehicle running attitude detection method of the present invention.

[0061] Figure 2 This is a schematic diagram showing the distribution of camera mounting points after further optimization of the rail vehicle running attitude detection method of the present invention.

[0062] Figure 3 This is a flowchart of the rail vehicle running attitude detection method of the present invention;

[0063] Figure 4 This is a schematic diagram of the operation of the detection device in the rail vehicle running attitude detection system of the present invention;

[0064] Figure 5 This is a schematic diagram of the triangular geometry of the optical imaging method for detecting the running attitude of a rail vehicle in the OXZ plane according to the present invention.

[0065] Figure 6 This is a schematic diagram of the triangular geometry of the optical imaging in the OYZ plane for the rail vehicle running attitude detection method of the present invention;

[0066] Figure 7 This is a photograph of the linear laser spot of the rail vehicle running attitude detection system of the present invention.

[0067] The attached figures are labeled as follows:

[0068] 100. Camera mounting point for the first inspection group; 200. Camera mounting point for the second inspection group; 300. Camera mounting point for the third inspection group; 1. Inspection box; 2. Camera bracket; 3. Area scan camera; 4. Mounting hole; 5. Laser emitter bracket; 6. Laser emitter; 7. Prism box bracket; 8. Prism box; 9. Transparent protective plate. Detailed Implementation

[0069] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0070] See Figure 3-7 In a preferred embodiment of the present invention, a method for detecting the running attitude of a rail vehicle is provided, comprising the following steps:

[0071] S1. Select three sets of measuring points at the position corresponding to the bottom of the train above the track. The first and second sets of measuring points are at the same end of the train, and the first and third sets of measuring points are on the same side of the train. Each set of measuring points includes two laser emission points and one camera mounting point. The two laser emission points continuously emit two parallel laser beams into the track. The two laser beams illuminate the track to form two linear laser spots. The camera mounting point is equipped with an area array camera that can continuously and clearly capture the linear laser spots. The area array camera uses a fixed-focus lens. The image plane of the area array camera is parallel to the track plane, and the focal points of the area array cameras in the three detection groups are not on a straight line.

[0072] S2. When the train is at the positive position 0 point, measure the distance between the two linear laser spots of the first to third sets of measuring points. The first set is 110 mm, the second set is 108 mm, and the third set is 110 mm.

[0073] S3. At time t during the train's operation, based on the principles of optical imaging and the properties of similar triangles, calculate the vertical and lateral dynamic offsets of the focal point of the area array camera relative to the reference position for each set of measuring points.

[0074] Establish a track-following coordinate system that moves forward with the train. The origin O is the projection of the geometric center point of the car body onto the track plane. The positive direction of the X-axis is the center line of the track plane along the direction of train travel. The positive direction of the Z-axis is the direction perpendicular to the track plane and upward. The Y-axis is perpendicular to the OXZ plane and points to the right.

[0075] Calculate the vertical dynamic offset ΔZ1 of the focal point of the first group of measuring points area array camera at time t, according to the formula:

[0076]

[0077] Where f is the focal length of the area array camera, and the focal length of the first group of measuring point area array cameras is 35 mm.

[0078] L AB The distance between linear laser spots on the track is 110 mm for the first group;

[0079] Let t be the distance between the two linear laser spots on the camera image at time t, and the distance of the first group is 7.4745 mm;

[0080] The distance between the two linear laser spots on the camera image when the train is in the positive position 0 point is 7.7000 mm;

[0081] Substitute into the calculation,

[0082]

[0083] Similarly, the vertical dynamic offsets ΔZ2 and ΔZ3 of the focal points of the second and third groups of measuring point area array cameras were calculated.

[0084] ΔZ2=0.1mm, ΔZ3=3.7mm.

[0085] Calculate the lateral dynamic offset ΔY1 of the focal point of the first group of measuring points area array camera at time t, according to the formula:

[0086]

[0087] Wherein, point A′ is the midpoint of the orbital plane corresponding to the linear laser spot a, and point C′ is the imaging point of point A′ on the image plane.

[0088] Let C′ be the distance between point C and point C at time t. The distance for the first group is -0.1815 mm.

[0089] The distance between point C′ and point C when the train is in its positive position 0 is 0.5610 mm.

[0090] L AB The distance between linear laser spots on the track is 110 mm for the first group;

[0091] Let t be the distance between the two linear laser spots on the camera image at time t, and the distance of the first group is 7.4745 mm;

[0092] The distance between the two linear laser spots on the camera image when the train is in the positive position 0 point is 7.7000 mm;

[0093] Substitute into the calculation,

[0094]

[0095] Similarly, the lateral dynamic offsets ΔY2 and ΔY3 of the focal points of the second and third groups of measuring point area array cameras were obtained through calculation.

[0096] ΔY2=-10.7mm, ΔY3=0.6mm.

[0097] S4. Calculate the five attitude parameters of the train at time t: roll angle α, pitch angle β, yaw angle γ, yaw offset Δy, and buoyancy offset Δz.

[0098] Calculate the roll angle α, pitch angle β, and yaw angle γ of the train at time t using the formula:

[0099]

[0100] Where W is the lateral distance between the camera mounting points of the two detection groups on the same head, which is directly measured using a millimeter ruler and is W = 1435 mm;

[0101] Substitute into the calculation,

[0102]

[0103] According to the formula:

[0104]

[0105] Where L is the longitudinal distance between the camera mounting points of the two detection groups on the same side, which is directly measured using a millimeter ruler and is found to be L = 13015 mm.

[0106] Substitute into the calculation,

[0107]

[0108] According to the formula:

[0109]

[0110] Substitute into the calculation,

[0111]

[0112] The formulas for calculating the yaw offset Δy and buoyancy offset Δz of the train at time t are as follows:

[0113]

[0114]

[0115] When i = 1, the sign before α is "+", the sign before β is "+", and the sign before γ is "+".

[0116] When i = 2, the sign before α is "-", the sign before β is "+", and the sign before γ is "+".

[0117] When i = 3, the sign before α is "+", the sign before β is "-", and the sign before γ is "-".

[0118] Let i = 1 and substitute it into the calculation.

[0119]

[0120]

[0121] S5. For different operating environments, set different weights for each attitude parameter, {p i, i∈(1,5)} are the weight parameters of five different attitude factors, S t Let S be the real-time weighted risk coefficient under the current operating environment, and S be the specified risk coefficient under the current operating environment. Then we have:

[0122] S t =p1|α|+p2|β|+p3|γ|+p4|Δy|+p5|Δz|

[0123] In a certain test, the operating environment was as follows: the maximum experimental speed was 160 km / h, the wind speed was 24.4 m / s, the route was straight with no slope, therefore, p1 = 0.2, p2 = p3 = 1, p4 = p5 = 0.02, and the specified S was 1. The MAX(S) measured during the monitoring period was... t The value is 0.36.

[0124] Compare S t Since S < 1, the train's operating posture is safe under this operating environment, and no warning is issued.

[0125] On the other hand, embodiments of the present invention also provide a rail vehicle running attitude detection system, comprising:

[0126] The detection device includes a detection box 1, see [link / reference]. Figure 4 The detection box 1 contains an area array camera 3, a laser emitter 6, and a prism box 8. The area array camera 3 is mounted on a camera bracket 2, the laser emitter 6 is mounted on a laser emitter bracket 5, and the prism box 8 is mounted on a prism box bracket 7. The detection box is fixed to the bottom of the train through mounting holes 4. The transparent protective plate 9 is made of transparent material. The protective plate 9 can prevent foreign objects such as wind and sand from entering the detection box 1 while ensuring clear imaging of the camera and laser, thus protecting the equipment inside. There are at least three sets of detection devices, which are respectively set at three non-collinear positions on the bottom of the train above the track. The first and second sets of detection devices are at the same end of the train, and the first and third sets of detection devices are on the same side of the train. The laser emitter continuously emits parallel lasers into the track, and the area array camera continuously captures the linear laser spots formed by the same set of lasers on the track and sends the data to the calculation and judgment device in real time.

[0127] The calculation and judgment device acquires data from the area array camera of the detection device, calculates the train's roll angle α, pitch angle β, yaw angle γ, yaw offset Δy, and buoyancy offset Δz at time t, and calculates the real-time weighted risk coefficient S under the current operating environment based on the five attitude parameters obtained. t By comparing S t Based on the risk coefficient S predetermined for the line under the current operating environment, determine whether the train's running posture is safe and send a signal to the signal output device;

[0128] A signal output device is used to receive instructions from the calculation and judgment device and issue a warning signal.

[0129] The above-mentioned rail vehicle running attitude detection system can implement various embodiments of the above-mentioned rail vehicle running attitude detection method and achieve the same beneficial effects, which will not be elaborated here.

[0130] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for detecting the running attitude of a rail vehicle, characterized in that, Includes the following steps: S1. Three detection groups, namely the first, second, and third, are set at the bottom of the same carriage of the train above the track. Each detection group includes a laser emission point and a camera mounting point. The laser emission point can continuously emit two parallel laser beams towards the track. The two laser beams illuminate the track surface to form two linear laser spots perpendicular to the center line of the track surface. The camera mounting point is equipped with an area array camera that can clearly capture the linear laser spots. The area array camera uses a fixed-focus lens. The image plane of the area array camera is parallel to the track plane, and the focal points of the area array cameras of the three detection groups are not on a straight line. S2. When the train is stationary on the track, it is considered to be in the positive position 0 point, that is, there is no offset on the track. At this time, the distance between the two linear laser spots of each detection group is measured, and the position of the focal point of the area array camera at this time is regarded as the reference position. S3. At time t during the train's operation, based on the principles of optical imaging and the properties of similar triangles, calculate the vertical dynamic offset ΔZ and lateral dynamic offset ΔY of the focal point of the area array camera relative to the reference position for each set of measuring points. S4. Calculate the five attitude parameters of the train at time t: roll angle α, pitch angle β, yaw angle γ, yaw offset Δy, and buoyancy offset Δz. S5, S t is the real-time weight risk coefficient under the current operating environment, and S is the pre-specified risk coefficient of the line under the current operating environment, which is calculated based on five attitude parameters to obtain S t , compare S t and S. When S t ≥ S, the running attitude of the train at time t in this operating environment is unsafe, and an early warning shall be issued; when S t < S, the running attitude of the train at time t in this operating environment is safe, and no early warning shall be issued.

2. The method for detecting the running attitude of a rail vehicle according to claim 1, characterized in that, In step S1, the plane where the camera mounting points of the three detection groups are located is parallel to the track plane.

3. The method for detecting the running attitude of a rail vehicle according to claim 2, characterized in that, In step S1, the camera mounting points of the first and second detection groups are located at the same end of the carriage, and their straight lines are perpendicular to the center line of the track; the camera mounting points of the first and third detection groups are located on the same side of the carriage, and their straight lines are parallel to the center line of the track.

4. The method for detecting the running attitude of a rail vehicle according to claim 3, characterized in that, In step S3, the step of calculating the vertical dynamic offset ΔZ includes: Establish a track-following coordinate system that moves forward with the train. The origin O is the projection of the geometric center point of the car body onto the track plane. The positive direction of the X-axis is the center line of the track plane along the direction of train travel. The positive direction of the Z-axis is the direction perpendicular to the track plane and upward. The Y-axis is perpendicular to the OXZ plane and points to the right. Calculate the vertical dynamic offset ΔZ1 of the focal point of the first detection group's area array camera at time t, where a and b are two linear laser spots in the first detection group; A and B are the projections of these two linear laser spots onto the OXZ plane, and the distance between them is denoted by L. AB L indicates AB The distance is equal to the distance between parallel linear laser spots a and b; points C and D are the corresponding imaging points of points A and B on the camera's image plane, and the distance between them is denoted by L. DC Indicates: F is the focal point of the area scan camera; MF and FH are the perpendicular segments from point F to lines AB and CD, respectively; L MF It is the camera's focal length f, L FH It is the distance from the focal point of the area array camera to the track plane. According to the principles of optical imaging and the properties of similar triangles, where: When the train is at the positive position 0, the apparent vertical offset is 0. It is the distance from the focal point of the area array camera to the track plane at this moment. The distance between points C and D at this moment is: At time t during the train's operation, It is the distance from the focal point of the area array camera to the track plane at this moment. The distance between points C and D at this moment is: The vertical dynamic offset ΔZ1 of the focal point of the first detection group's area array camera is: Similarly, the vertical dynamic offsets ΔZ2 and ΔZ3 of the focal points of the area array cameras in the second and third detection groups were calculated.

5. The method for detecting the running attitude of a rail vehicle according to claim 4, characterized in that, In step S3, the step of calculating the lateral dynamic offset ΔY includes: Calculate the lateral dynamic offset ΔY1 of the focal point of the first detection group area array camera at time t, where point A' is the midpoint of the track plane and point C' is the imaging point of point A' on the image plane, using L... C'C The distance between point C and point C' is represented by L. A'A Let A' represent the distance between point A and point A, then: When the train is at the positive position 0, the lateral dynamic offset ΔY1 is considered to be 0, and this is used as the measurement standard for the lateral dynamic offset at time t. This is the distance between point A' and point A at this moment. This is the distance between point C and point C' at this moment, where: At time t during the train's operation, This is the distance between point A' and point A at this moment. The distance between point C and point C' at this moment is: The lateral dynamic offset ΔY1 of the focal point of the first detection group's area array camera is: Similarly, the lateral dynamic offsets ΔY2 and ΔY3 of the focal points of the second and third detection group area array cameras were calculated.

6. The method for detecting the running attitude of a rail vehicle according to claim 5, characterized in that, In step S4, the steps of calculating the roll angle α, pitch angle β, and yaw angle γ of the train at time t include: W represents the lateral distance between the area scan cameras of the two detection groups on the same head, i.e., the distance between the area scan cameras along the Y direction, which is directly measured. The roll angle α is: L is the longitudinal distance between the area scan cameras of the two detection groups on the same side, that is, the distance between the area scan cameras along the X direction, which is directly measured. The nod angle β is: The head-shaking angle γ is:

7. The method for detecting the running attitude of a rail vehicle according to claim 6, characterized in that, In step S4, the formulas for calculating the yaw offset Δy and buoyancy offset Δz of the train at time t are as follows: When i = 1, the sign before α is "+", the sign before β is "+", and the sign before γ is "+". When i = 2, the sign before α is "-", the sign before β is "+", and the sign before γ is "+". When i = 3, the sign before α is "+", the sign before β is "-", and the sign before γ is "-".

8. The method for detecting the running attitude of a rail vehicle according to claim 1, characterized in that, In step S5, S t The calculation formula is as follows: S t =p1|α|+p2|B|+p3|γ|+p4|Δy|+p5|Δz| Among them, different weights are set for each attitude parameter for different operating environments, {p i ,i∈(1,5)} are the weight parameters of five different attitude factors.

9. A rail vehicle running attitude detection system applied to the method of any one of claims 1-8, characterized in that, include: The detection device comprises at least three sets, each set including a laser emitter, a prism box, and an area array camera. The laser emitter is capable of continuously emitting two parallel laser beams toward the track, which illuminate the track to form a linear laser spot. The area array camera is capable of continuously capturing the linear laser spot of the same set of detection devices and sending the data to the calculation and judgment device in real time after obtaining a clear image. The calculation and judgment device acquires data from the area array camera of the detection device, calculates the roll angle α, pitch angle β, yaw angle γ, yaw offset Δy, and buoyancy offset Δz of the train at time t, and determines whether the train's running posture is safe, and sends a signal to the signal output device. A signal output device is used to receive instructions from the calculation and judgment device and issue a warning signal.

10. The rail vehicle running attitude detection system according to claim 9, characterized in that, The detection device includes a detection box (1), which includes an area array camera (3), a laser emitter (6), and a prism box (8). The laser emitted by the laser emitter (6) passes through the prism box (8) and is converted into two parallel laser beams.

Images