CN104634298A - Existing railway measuring method based on LIDAR (Light Detection and Ranging) track point cloud data - Google Patents

Abstract

The invention relates to an existing railway measuring method based on LIDAR (Light Detection and Ranging) track point cloud data. The existing railway line measurement has the defects of poor flexibility and limited measuring length. The method comprises the following steps: recording continuous three-dimensional coordinate data in a track running process by using a vehicle-mounted LIDAR measurement instrument for serving as track point cloud data; extracting a center line between two tracks, translating o obtain a route center line, and calculating the fixed-length chord slope of the route center line: evaluating the slope by taking new data serving as a point set as a difference, generating a fixed-length chord slope sequence, and filtering; picking segmentation points on the basis of the smooth fixed-length chord slope, and decomposing a continuous route into curve units; evaluating the intersection points among the curves according to front and back intermediate straight lines of each unit, fitting the radius and circle center of a circular curve part of the route, and resolving a corresponding route parameter and mileage. A basic unit set of route curve elements is constructed according to the LIDAR track point cloud data, so that the aims of route reconstruction and retesting are fulfilled, high flexibility, high measuring efficiency and a simple process are realized, and measured data is accurate and reliable.

Description

Based on the existing railway survey method of LIDAR track cloud data

Technical field

The invention belongs to railroad survey technical field, be specifically related to a kind of existing railway survey method based on LIDAR track cloud data.

Background technology

Railway track (comprising rail and roadbed) is the most important infrastructure of railway operation, and track, directly by the immense pressure that car wheel-set transmits, causes it to produce various distortion, and the high-speed smooth affecting train is run.The distortion of track adopts irregularity to be described.Track irregularity refers to the geometric configuration of track, and size and relative space position are relative to the deviation of its normal state.

The prerequisite of the ride comfort of assessment circuit is the locus state obtaining circuit, and this just requires to carry out repetition measurement to existing railway circuit.Traditional plane curve measuring method based on theory of involute, in practical operation, on curve, the undulate quantity error of measuring point is comparatively large, is substantially eliminated.Measure the planimetric position of orbit centre and the coordinate method of tread elevation based on total powerstation and spirit-leveling instrument, become the main measurement method of railway both wired repetition measurement at present.But total powerstation Arrange mirror in any point method exists two problems: one is be difficult to determine the practical center between two tracks, need to measure with steel ruler or rail chi, such method is time-consuming, effort, and measuring accuracy is lower, and cost is higher.Two is that total powerstation needs to be placed in the visible scope of whole section, and under the limited condition of mirror condition is not had an X-rayed or put to circuit, Arrange mirror in any point needs to arrange a lot of turning points, and computation process is comparatively loaded down with trivial details.And larger measuring error can be caused.Then cause the method to be mainly used in the measurement of single curve, be difficult to the continuous coverage being implemented to circuit.Therefore, pendulous frequency is few, and computation complexity is low, measure and adjustable lengthly have important Research Significance and using value based on track cloud data existing railway survey method.

LIDAR and Light Detection And Ranging, Chinese is laser radar, is the multiple sophisticated technology spatial measurement system such as the range finding of a kind of set laser, digital airborne photography, GPS (GPS) and inertial navigation system (INS).In recent years, its ability obtaining three-dimensional data is fast by extensive concern.

Summary of the invention

The object of this invention is to provide a kind of existing railway survey method based on LIDAR track cloud data, overcome existing survey of existing railway very flexible, measure the more problem of length limited, operating personnel's upper track operations number.

The technical solution adopted in the present invention is:

Based on the existing railway survey method of LIDAR track cloud data, it is characterized in that:

Described track point cloud, for vehicle-mounted LIDAR measuring instrument is along the continuous three-dimensional coordinate data recorded in rail running process, comprises left rail and right rail data;

Measuring method specifically comprises the following steps:

Step one: left and right rail, respectively according to certain length, extracts the center line of two rails;

Step 2: translation left and right rail center line obtains central lines of track, to move to right 717.5mm by left rail center line, and right rail center line moves to left 717.5mm;

Step 3: the fixed length string slope of computational scheme center line:

With the new data point set that step 2 produces, asking slope by making difference between any two, producing fixed length string slope sequence, and adopting rlowess method to carry out filtering process, staging treating is carried out to continuous curve;

Step 4: on the basis of level and smooth fixed length string slope, picks up corresponding waypoint, continuous print circuit is resolved into several curved unit;

Step 5: the front and back intermediate straight line of each unit drawn according to step 4 obtains the intersection point JD of curve, family of circles method is adopted to expand radius gradually, gone out radius and the center of circle of circuit circular curve part by least square fitting, thus calculate corresponding line parameter circuit value and mileage.

In step one:

Left and right rail, respectively according to certain length, is concentrated order to extract n data from original point, is extracted the center line of two rails according to least square method;

$Q\≡Q(a,b)=\underset{i-1}{\overset{n}{\mathrm{\Σ}}}{e_i}^2=\underset{i-1}{\overset{n}{\mathrm{\Σ}}}{(y_i-{\hat{y}}_i)}^2=\underset{i-1}{\overset{n}{\mathrm{\Σ}}}{(y_i-a-b{x}_i)}^2$

In formula, n represents the point participating in matching, take 6m as default sample interval, calculates within the scope of 1m comprise 300 three-dimensional point data, i.e. n=1800 according to the parameter of measuring instrument.

In step 2, obtain central lines of track according to following rule:

$\left\{\begin{array}{c}{x}^{\′}=x+d\left|\sin \mathrm{\θ}\right|v\\ y^{\′}=y+d\left|\sin \mathrm{\θ}\right|v\end{array}\right.$

In formula:

d＝717.5mm；

θ is the angle of this section of straight line and X-direction;

The value of v is as shown in the table:

。

In step 3, carry out filtering process according to following rule:

$w_i=\left\{\begin{array}{cc}{(1-(r_i/6\mathrm{MAD}{)}^2))}^2,& \left|\mathrm{ri}\right|<6\mathrm{MAD},\\ 0,& \left|\mathrm{ri}\right|\&GreaterEqual;6\mathrm{MAD},\end{array}\right.$

In formula:

W _irepresent the weights of each point in subrange;

MAD is the absolute dispersion in subrange;

R _irepresent the fixed length slope value of each point in subrange;

I represents the digit of fixed length slope sequence.

In step 3, described fixed length string slope sequence, compared to the few point of former data set, after segmentation completes, moves one by fragment bit postpone.

In step 5, go out after circuit circular curve part obtains curve element by least square fitting, under curvilinear coordinate system, calculated curve unique point coordinate:

ZH point: X _zh=X ₀-m ₁

Y _zh＝0

HY point: X _hy=X _zh+ m ₁+ R _ssin β ₀

Y _hy＝R _s+P ₁-R _scosβ ₀

QZ point: $X_{\mathrm{qz}}=X_{\mathrm{zh}}+m_1+R_s\sin \frac{\mathrm{\&alpha;}}{2}$

$Y_{\mathrm{qz}}=R_s+P_1-R_s\cos \frac{\mathrm{\&alpha;}}{2}$

YH point: X _yh=X _zh+ m ₁+ R _ssin (alpha-beta ₀)

Y _yh＝R _s+P ₁-R _scos(α-β ₀)

HZ point: X _hz=X _jd+ T ₂cos α

Y _hz＝Y _jd+T ₂sinα

In formula:

β ₀for the grazing angle of adjustment curve, namely the tangent line put of HY (or YH) and ZH (or HZ) put the angle of cut of tangent line, that is the central angle corresponding to the extended line part of circular curve one end, (X _jd, Y _jd) be intersecting point coordinate, i.e. (((R+P) × sec α/2 × sin α+m), 0).

In step 5, after obtaining line parameter circuit value, each coordinate points in loop iteration circuit, obtains corresponding curve element point coordinate and corresponding mileage number, and carries out compensating computation according to known accurate coordinates to the result calculated.

The present invention has the following advantages:

A kind of existing railway survey method based on LIDAR track cloud data that the present invention proposes, utilize vehicle-mounted LIDAR technology, image data, by to data processing, extract track centerline and carry out the plane of Existing Railway Line and the segmentation of vertical section and reconstruct, be applicable to the track data calculating random length, the length of circuit is unrestricted.The method has simple and convenient ageing, and measuring process does not need human intervention or only comprises few human assistance.

Accompanying drawing explanation

Fig. 1 is the process flow diagram that the present invention is based on track cloud data existing railway survey method.

Fig. 2 is the schematic diagram of left and right of the present invention track cloud data computational scheme center line.

Fig. 3 is this track circuit center line fixed length string slope variation schematic diagram.

Fig. 4 is the schematic diagram of circuit a certain unit flat section algorithm.

Embodiment

Below in conjunction with embodiment, the present invention will be described in detail.

The existing railway survey method based on LIDAR track cloud data that the present invention relates to, it is a kind of existing railway repetition measurement method based on track cloud data, described track point cloud is for vehicle-mounted LIDAR measuring instrument is along the continuous three-dimensional coordinate data recorded in rail running process, comprise left rail and right rail data, measuring method specifically comprises the following steps:

Step one: left and right rail, respectively according to certain length, extracts the center line of two rails.

Left and right rail, respectively according to certain length, is concentrated order to extract n data from original point, is extracted the center line of two rails according to least square method;

$Q\&equiv;Q(a,b)=\underset{i-1}{\overset{n}{\mathrm{\&Sigma;}}}{e_i}^2=\underset{i-1}{\overset{n}{\mathrm{\&Sigma;}}}{(y_i-{\hat{y}}_i)}^2=\underset{i-1}{\overset{n}{\mathrm{\&Sigma;}}}{(y_i-a-b{x}_i)}^2$

In formula, n represents the point participating in matching, take 6m as default sample interval, calculates within the scope of 1m comprise 300 three-dimensional point data, i.e. n=1800 according to the parameter of measuring instrument.

Step 2: translation left and right rail center line obtains central lines of track, to move to right 717.5mm by left rail center line, and right rail center line moves to left 717.5mm.

Central lines of track is obtained according to following rule:

$\left\{\begin{array}{c}{x}^{\&prime;}=x+d\left|\sin \mathrm{\&theta;}\right|v\\ y^{\&prime;}=y+d\left|\sin \mathrm{\&theta;}\right|v\end{array}\right.$

In formula:

d＝717.5mm；

θ is the angle of this section of straight line and X-direction;

The value of v is as shown in the table:

。

Step 3: the fixed length string slope of computational scheme center line:

With the new data point set that step 2 produces, slope is asked by making difference between any two, produce fixed length string slope sequence, and adopt rlowess method (the local weighted moving least squares method of sane type) to carry out filtering process, to carry out staging treating to continuous curve.

Filtering process is carried out according to following rule:

$w_i=\left\{\begin{array}{cc}{(1-(r_i/6\mathrm{MAD}{)}^2))}^2,& \left|\mathrm{ri}\right|<6\mathrm{MAD},\\ 0,& \left|\mathrm{ri}\right|\&GreaterEqual;6\mathrm{MAD},\end{array}\right.$

In formula:

W _irepresent the weights of each point in subrange;

MAD is the absolute dispersion in subrange;

R _irepresent the fixed length slope value of each point in subrange;

I represents the digit of fixed length slope sequence.

Described fixed length string slope sequence, compared to the few point of former data set, after segmentation completes, moves one by fragment bit postpone.

Step 4: on the basis of level and smooth fixed length string slope, picks up corresponding waypoint, continuous print circuit is resolved into several curved unit.

According to fixed length string slope---period figure, the place period of intermediate straight line can be picked out easily, thus continuous print circuit is resolved into several curved unit.

Step 5: the front and back intermediate straight line of each unit drawn according to step 4 obtains the intersection point JD of curve, family of circles method is adopted to expand radius gradually, gone out radius and the center of circle of circuit circular curve part by least square fitting, thus calculate corresponding line parameter circuit value and mileage.

Go out after circuit circular curve part obtains curve element by least square fitting, under curvilinear coordinate system, calculated curve unique point coordinate:

ZH point: X _zh=X ₀-m ₁

Y _zh＝0

HY point: X _hy=X _zh+ m ₁+ R _ssin β ₀

Y _hy＝R _s+P ₁-R _scosβ ₀

QZ point: $X_{\mathrm{qz}}=X_{\mathrm{zh}}+m_1+R_s\sin \frac{\mathrm{\&alpha;}}{2}$

$Y_{\mathrm{qz}}=R_s+P_1-R_s\cos \frac{\mathrm{\&alpha;}}{2}$

YH point: X _yh=X _zh+ m ₁+ R _ssin (alpha-beta ₀)

Y _yh＝R _s+P ₁-R _scos(α-β ₀)

HZ point: X _hz=X _jd+ T ₂cos α

Y _hz＝Y _jd+T ₂sinα

In formula:

β ₀for the grazing angle of adjustment curve, namely the tangent line put of HY (or YH) and ZH (or HZ) put the angle of cut of tangent line, that is the central angle corresponding to the extended line part of circular curve one end, (X _jd, Y _jd) be intersecting point coordinate, i.e. (((R+P) × sec α/2 × sin α+m), 0).After obtaining line parameter circuit value, each coordinate points in loop iteration circuit, obtains corresponding curve element point coordinate and corresponding mileage number, and carries out compensating computation according to known accurate coordinates to the result calculated.

The inventive method is not limited to both wired cloud data of vehicle-mounted LIDAR measuring instrument, is also applicable to adopt any-mode to obtain both wired repetition measurement of the three-dimensional coordinate data of track circuit.Content of the present invention is not limited to cited by embodiment, and the conversion of those of ordinary skill in the art by reading instructions of the present invention to any equivalence that technical solution of the present invention is taked, is claim of the present invention and contains.

Claims (7)

1., based on the existing railway survey method of LIDAR track cloud data, it is characterized in that:

Described track point cloud, for vehicle-mounted LIDAR measuring instrument is along the continuous three-dimensional coordinate data recorded in rail running process, comprises left rail and right rail data;

Measuring method specifically comprises the following steps:

Step one: left and right rail, respectively according to certain length, extracts the center line of two rails;

Step 2: translation left and right rail center line obtains central lines of track, to move to right 717.5mm by left rail center line, and right rail center line moves to left 717.5mm;

Step 3: the fixed length string slope of computational scheme center line:

With the new data point set that step 2 produces, asking slope by making difference between any two, producing fixed length string slope sequence, and adopting rlowess method to carry out filtering process, staging treating is carried out to continuous curve;

Step 4: on the basis of level and smooth fixed length string slope, picks up corresponding waypoint, continuous print circuit is resolved into several curved unit;

Step 5: the front and back intermediate straight line of each unit drawn according to step 4 obtains the intersection point JD of curve, family of circles method is adopted to expand radius gradually, gone out radius and the center of circle of circuit circular curve part by least square fitting, thus calculate corresponding line parameter circuit value and mileage.

2. the existing railway survey method based on LIDAR track cloud data according to claim 1, is characterized in that:

In step one:

Left and right rail, respectively according to certain length, is concentrated order to extract n data from original point, is extracted the center line of two rails according to least square method;

$Q\&equiv;Q(a,b)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{e_i}^2=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{(y_i-{\hat{y}}_i)}^2=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{(y_i-a-b{x}_i)}^2$

In formula, n represents the point participating in matching, take 6m as default sample interval, calculates within the scope of 1m comprise 300 three-dimensional point data, i.e. n=1800 according to the parameter of measuring instrument.

3. the existing railway survey method based on LIDAR track cloud data according to claim 2, is characterized in that:

In step 2, obtain central lines of track according to following rule:

$\left\{\begin{array}{c}{x}^{\&prime;}=x+d\left|\sin \mathrm{\&theta;}\right|v\\ y^{\&prime;}=y+d\left|\sin \mathrm{\&theta;}\right|v\end{array}\right.$

In formula:

d＝717.5mm；

θ is the angle of this section of straight line and X-direction;

The value of v is as shown in the table:

4. the existing railway survey method based on LIDAR track cloud data according to claim 3, is characterized in that:

In step 3, carry out filtering process according to following rule:

$w_i=\left\{\begin{array}{cc}{(1-{(r_i/6\mathrm{MAD})}^2))}^2,& \left|\mathrm{ri}\right|<6\mathrm{MAD},\\ 0,& \left|\mathrm{ri}\right|6\mathrm{MAD},\end{array}\right.$

In formula:

W _irepresent the weights of each point in subrange;

MAD is the absolute dispersion in subrange;

R _irepresent the fixed length slope value of each point in subrange;

I represents the digit of fixed length slope sequence.

5. the existing railway survey method based on LIDAR track cloud data according to claim 4, is characterized in that:

In step 3, described fixed length string slope sequence, compared to the few point of former data set, after segmentation completes, moves one by fragment bit postpone.

6. the existing railway survey method based on LIDAR track cloud data according to claim 5, is characterized in that:

In step 5, go out after circuit circular curve part obtains curve element by least square fitting, under curvilinear coordinate system, calculated curve unique point coordinate:

ZH point: $\begin{array}{c}{X}_{\mathrm{zh}}=X_0-m_1\\ Y_{\mathrm{zh}}=0\end{array}$

HY point: $\begin{array}{c}{X}_{\mathrm{hy}}=X_{\mathrm{zh}}+m_1+R_s\sin {\mathrm{\&beta;}}_0\\ X_{\mathrm{hy}}=R_s+P_1-R_s\cos {\mathrm{\&beta;}}_0\end{array}$

QZ point: $\begin{array}{c}{X}_{\mathrm{qz}}=X_{\mathrm{zh}}+m_1+R_s\sin \frac{\mathrm{\&alpha;}}{2}\\ X_{\mathrm{qz}}=R_s+P_1-R_s\cos \frac{\mathrm{\&alpha;}}{2}\end{array}$

YH point: $\begin{array}{c}{X}_{\mathrm{yh}}=X_{\mathrm{zh}}+m_1+R_s\sin (\mathrm{\&alpha;}-{\mathrm{\&beta;}}_0)\\ X_{\mathrm{yh}}=R_s+P_1-R_s\cos (\mathrm{\&alpha;}-{\mathrm{\&beta;}}_0)\end{array}$

HZ point: $\begin{array}{c}{X}_{\mathrm{hz}}=X_{\mathrm{jd}}+T_2\cos \mathrm{\&alpha;}\\ X_{\mathrm{hz}}=Y_{\mathrm{jd}}+T_2\sin \mathrm{\&alpha;}\end{array}$

In formula:

β ₀for the grazing angle of adjustment curve, namely the tangent line put of HY (or YH) and ZH (or HZ) put the angle of cut of tangent line, that is the central angle corresponding to the extended line part of circular curve one end, (X _jd,y _jd) be intersecting point coordinate, i.e. (((R+P) sec α/2sin α+m), 0).

7. the existing railway survey method based on LIDAR track cloud data according to claim 6, is characterized in that:

In step 5, after obtaining line parameter circuit value, each coordinate points in loop iteration circuit, obtains corresponding curve element point coordinate and corresponding mileage number, and carries out compensating computation according to known accurate coordinates to the result calculated.