Differential mapping and orbit mapping method with weak inertial navigation dependence

Abstract

The invention discloses a difference mapping and track mapping method with weak inertial navigation dependence. The method comprises the following steps: S1, data acquisition: acquiring and obtaining point cloud data around a track by using a data acquisition system; s2, defining a system coordinate system L and a world coordinate system W; s3, track point cloud mileage estimation: point cloud re-projection; extracting local feature descriptors; calculating a pose transformation matrix # imgabs0 #, calculating a pose transformation matrix # imgabs1 #, and converging the pose transformation matrix # imgabs2 # and the pose transformation matrix # imgabs3 # to obtain a final pose transformation matrix; carrying out iterative convergence solution; s4, mapping orbits; S41, unifying a world coordinate system W; and S42, performing track prior mapping. The method has the advantages of being high in sparse tolerance, low in drift, low in complexity and high in detail reduction degree, the dependence of auxiliary hardware is greatly reduced, the track scene reconstruction precision is higher, and the economical efficiency and popularization and application prospects are better.

Description

Technical Field

[0001] The invention belongs to the technical field of high-precision railway measurement, and in particular relates to a difference mapping and track mapping method with weak inertial navigation dependence. Background Art

[0002] In the railway operation and maintenance stage, obtaining high-precision scene data and corresponding characteristic geometric parameters is an important data input for railway status detection and monitoring, maintenance and remediation evaluation, and operational safety maintenance. At present, my country is in a period of rapid development of intelligent equipment replacing manual labor, especially more mature track inspection instruments, mobile 3D scanning systems and other equipment, which shine in the field of high-precision railway measurement. However, in actual applications, in areas such as tunnels, underground, and blind spots, satellite positioning and navigation signals may be interfered with, lost, or have no signal. It is usually necessary to add high-precision inertial navigation as a data source and correction in the mileage direction, but this method mainly has the following problems:

[0003] (1) The accumulation of inertial navigation errors causes the mileage data obtained during the measurement process to gradually deviate from the true value. However, track-related application scenarios are non-closed-loop conditions, and it is impossible to rely on traditional models to eliminate or weaken errors, resulting in large measurement data errors and unable to meet the high-precision requirements of engineering applications.

[0004] (2) The mobile data collection process mainly relies on the integration of small incremental movements that change over time, which will cause data drift, resulting in holes or distortion in the point cloud results, directly affecting the authenticity of track measurement and endangering driving safety;

[0005] (3) The pose estimation process mainly relies on the inertial navigation output results to compensate the estimated data. For acquisition devices with more than two degrees of freedom, such as lidar, there are problems such as regional boundary restrictions, missing dead zones and field of view distortion. The accuracy, precision and confidence of the original data are low.

[0006] In view of the above problems and weaknesses, it is urgent to develop a difference mapping and orbit mapping method with weak inertial navigation dependence. Summary of the invention

[0007] In order to solve the problems existing in the prior art, the present invention provides a method for weak inertial navigation dependency difference mapping and orbit mapping with high sparse tolerance, low drift and low complexity. The method of the present invention realizes accurate estimation of speed, position and attitude and fine registration mapping of local point clouds, thereby reducing the dependence on positioning and inertial measurement technology, and greatly reducing hardware costs and application limitations.

[0008] The technical scheme and steps of the present invention are as follows:

[0009] S1, data collection: Use the data collection system to collect and obtain point cloud data around the track, and record the point cloud data obtained by the k-th scan of the data collection system as point cloud P k , and add the point cloud data obtained by each scan to the set P. Then the data acquisition system performs N scans to obtain the set P = {P1, P2, ..., P k-1 ,P k ,P k+1 ,…P N};

[0010] The value range of k is k=1,2…N, where N is the total number of continuous scans performed by the data acquisition system; point cloud P k Satisfy the coplanar geometric relationship;

[0011] S2, define the system coordinate system L and the world coordinate system W:

[0012] System coordinate system L: The geometric center position of the laser radar in the data acquisition system is the origin position of the system coordinate system L. The positive direction of the z-axis of the system coordinate system L is the same as the direction of the initial posture of the data acquisition system. The y-axis of the system coordinate system L is perpendicular to the z-axis and the positive direction of the y-axis is away from the center of the earth. The x-axis, y-axis and z-axis of the system coordinate system L form a right-handed coordinate system.

[0013] Among them, as the data acquisition system moves, the corresponding system coordinate system needs to be constructed at each scan, and the system coordinate system constructed by the kth scan is recorded as L k ; The point cloud P k Any point A in i In the system coordinate system L k The following is expressed as i=1,2,…,M, where M is the point cloud P k The total number of inliers;

[0014] World coordinate system W: The world coordinate system W completely coincides with the system coordinate system L1 constructed during the first scan of the data acquisition system, where the point cloud P k Any point A in i In the world coordinate system W k The following is expressed as

[0015] S3, track point cloud mileage estimation, specifically includes the following steps:

[0016] S31, point cloud reprojection: Set P = {P1, P2, ..., P k-1 ,P k ,P k+1 ,…P N Each point cloud in} is reprojected to its corresponding scanning end time to obtain the set

[0017] Among them, the point cloud P k The scan start time is t k , point cloud P k The scan end time is t k+1 , the point cloud P k Reproject to t k+1 Timestamp to get point cloud and add to the collection Therefore, the start time to the end time of the k+1th scan of the data acquisition system includes the point cloud and P k+1 ;

[0018] S32, extract local feature descriptors and obtain point cloud P according to the registration mapping relationship k+1 The boundary point set φ in k+1 and smooth point set Ω k+1 , point cloud The boundary point set in and smooth point set

[0019] S321, point cloud P in set P k Select a feature point A j As the initial reference point, the set P contains the initial reference point A j All point clouds are added to the set S1 = {P k ,…,P l}, in the point cloud P of the set P l Select a feature point A j ' as the reference point, and the set P contains the reference point A j All point clouds of ′ are added to the set S2 = {P l ,…,P n Repeat the above process in} until all point clouds in set P are divided and sets S1, S2…S are obtained. c ;

[0020] Among them, the value ranges of k, l, n and c are all 1, 2…N, and l>k, n>l;

[0021] S322, calculate the point cloud P in the set S1 k Except for the initial reference point A j In addition, A j is the local feature descriptor value η of each point of the initial reference point; k The point with the largest local feature descriptor value is taken as the boundary point to be selected, and the point with the smallest local feature descriptor value is taken as the smoothing point to be selected;

[0022] From the point cloud Pk and point cloud P k+1 The registration mapping relationship between them is used to obtain the point cloud P k+1 The boundary point set φ in k+1 and smooth point set Ω k+1 ; From point cloud With point cloud P k+1 The registration mapping relationship between them is used to obtain the point cloud The boundary point set in and smooth point set

[0023] Among them, the calculation point cloud P k Except for the initial reference point, each point is A j The calculation formula for the local feature descriptor value η of the initial reference point is as follows:

[0024]

[0025] Among them, A j ,A i ∈S,j≠i, M is the point cloud P k The total number of inliers;

[0026] S323, if the number of boundary points to be selected or the number of smooth points to be selected is zero or M, re-execute S321-S322, that is, re-select the initial reference point and recalculate; otherwise, execute S33;

[0027] S33, calculate the pose transformation matrix The specific steps are as follows:

[0028] S331, define the pose transformation matrix as At the current timestamp t k+2 Next, let [t k+1 ,t k+2 ] The pose transformation matrix of the data acquisition system during the time period is: Among them, t x ,t y ,t z Respectively represent the translation transformation of the x, y, and z axes in the L coordinate system, θ x ,θ y ,θ z Respectively represent the rotation transformation of the x, y, and z axes in the L coordinate system;

[0029] S332, by combining formula (2) and formula (3), the pose transformation matrix can be solved as follows:

[0030]

[0031] in, is the point cloud Pk+1 The boundary point set Φ k+1 Or smooth point set Ω k+1 The pose transformation matrix of any point q in t q Indicates the timestamp of point q, is the point cloud P in the L coordinate system k The coordinates of any point q in the function σ(*,n,m) are the values ​​of the nth to mth items of * ( Representation Matrix The corresponding homogeneous linear equation system consists of the solutions of the nth to mth terms), and the Rodrigues formula of R is as follows:

[0032] R=e ωθ =I+ωsinθ+ω 2 (1-cosθ)

[0033]

[0034]

[0035] Among them, the function Ψ(*) is a skew-symmetric matrix taking the variable *, and I is the identity matrix;

[0036] S34, the boundary point set Φ in S33 k+1 and smooth point set Ω k+1 Replace with boundary point set and smooth point set Execute step S33 to calculate the corresponding posture transformation matrix The pose transformation matrix and the pose transformation matrix Converge to get the final pose transformation matrix;

[0037] S35, iterative convergence solution: traverse the set S2…S c , execute S322-S34 to obtain the final posture transformation matrix corresponding to each set, and then obtain the spatial motion estimation curve of the data acquisition system corresponding to the set P, wherein the abscissa of the spatial motion estimation curve is the acquisition timestamp, and the ordinate is the final posture transformation matrix in S34;

[0038] S36, repeat S1-S35 until the point cloud P obtained by the last scan of the data acquisition system is obtained. final The corresponding spatial motion estimation curve of the data acquisition system;

[0039] S4, mapping track map:

[0040] S41, unification of the world coordinate system W: In the spatial motion estimation curve obtained in step S36, each time stamp includes corresponding point cloud data containing posture information, and all point cloud data are registered into the world coordinate system W using the real space mapping relationship;

[0041] S42, orbital prior mapping:

[0042] According to the prior feature information in the track application scenario, the point cloud fine registration algorithm is used to fuse the point clouds of adjacent timestamps in the point cloud under the world coordinate system W to realize track mapping.

[0043] In S1, the data acquisition system uses a self-mobile RC-SLAM high-precision three-dimensional scanning system, which can also be replaced by other devices with the same functions and the replacement equipment at least includes a calibrated dual-axis laser radar.

[0044] In S321, in the point cloud P l The method for selecting the reference point is: l After Delaunay triangulation, the average Euclidean distance d is obtained. l Select the initial reference point A j The Euclidean distance of The feature point A j ' as the reference point, wherein the initial reference point A j The reference points are all points outside the plane that are approximately parallel to the laser projection direction when the data acquisition system is scanning.

[0045] In S42, the priori characteristic information includes spatial seamless continuity and local consistency of rail profile, and within a certain time period, it is considered that the weight of logical information is higher than the weight of measurement information.

[0046] The updating frequency of S41 should be 10 times higher than that of S42, and the field of view overlap during the updating process should be higher than 70%.

[0047] In S322, n=1, m=3.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. The difference mapping and orbit mapping method of the present invention adopts a point cloud data classification and coordinate system registration method based on local feature descriptors, which greatly reduces the dependence on hardware parameters of auxiliary equipment such as inertial navigation, IMU, and gyroscope, and has higher economy and popularization and applicability. At the same time, it achieves better theoretical mapping accuracy with higher computing power.

[0050] 2. The difference mapping and orbit mapping method of the present invention adopts parallel calculation of the world coordinate system and orbit prior mapping mapping, which ensures the feasibility of convergence of problems in practical engineering applications. At the same time, the difference dual-frequency mapping process ensures the accuracy and restoration of local details, and weakens the cumulative error in practical applications.

[0051] 3. The local feature descriptor used in the difference mapping and track mapping method of the present invention can more effectively extract the local details and geometric features of the point cloud, further improve the accuracy of the point cloud fine registration results, and achieve high-precision reconstruction of the track scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a flow chart of the present invention;

[0053] Figure 2 It is a schematic diagram of point cloud data projection of the present invention;

[0054] Figure 3 It is a schematic diagram of difference mapping and track mapping of the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0056] A difference mapping and orbit mapping method with weak inertial navigation dependence, the process is as follows Figure 1 As shown, the following steps are included:

[0057] S1, data collection: Use the data collection system to collect and obtain point cloud data around the track, and record the point cloud data obtained by the k-th scan of the data collection system as point cloud P k , and add the point cloud data obtained by each scan to the set P. Then the data acquisition system performs N scans to obtain the set P = {P1, P2, ..., P k-1 ,P k ,P k+1 ,…P N}.

[0058] The data acquisition system uses a self-mobile RC-SLAM high-precision three-dimensional scanning system, which can also be replaced by other devices with the same function and the replacement device at least includes a calibrated dual-axis laser radar; the value range of k is k = 1, 2...N, N is the total number of continuous scans performed by the data acquisition system; the point cloud P k Satisfy the coplanar geometric relationship.

[0059] S2, define the system coordinate system L and the world coordinate system W:

[0060] System coordinate system L: The geometric center position of the laser radar in the data acquisition system is the origin position of the system coordinate system L. The positive direction of the z-axis of the system coordinate system L is the same as the direction of the initial posture of the data acquisition system. The y-axis of the system coordinate system L is perpendicular to the z-axis and the positive direction of the y-axis is away from the center of the earth. The x-axis, y-axis and z-axis of the system coordinate system L form a right-handed coordinate system.

[0061] Among them, as the data acquisition system moves, the corresponding system coordinate system needs to be constructed at each scan, and the system coordinate system constructed by the kth scan is recorded as L k ; The point cloud P k Any point A in i In the system coordinate system L k The coordinates below are i=1,2,…,M, where M is the point cloud P k The total number of inliers.

[0062] World coordinate system W: The world coordinate system W completely coincides with the system coordinate system L1 constructed during the first scan of the data acquisition system, where the point cloud P k Any point A in i In the world coordinate system W k The coordinates below are

[0063] S3, track point cloud mileage estimation, specifically includes the following steps:

[0064] S31, point cloud reprojection: Set P = {P1, P2, ..., P k-1 ,P k ,P k+1 ,…P N Each point cloud in} is reprojected to its corresponding scanning end time to obtain the set

[0065] Among them, Figure 2 As shown, the point cloud P k The scan start time is t k , point cloud P k The scan end time is t k+1 , the point cloud P k Reproject to t k+1 Timestamp to get point cloud and add to the collection Therefore, the start time to the end time of the k+1th scan of the data acquisition system includes the point cloud and P k+1 .

[0066] S32, extract local feature descriptors:

[0067] S321, point cloud P in set P k Select a feature point A j As the initial reference point, the set P contains the initial reference point A j All point clouds are added to the set S1 = {P k ,…,P l}; in the point cloud P of the set P l Select a feature point A j ' as the reference point, and the set P contains the reference point A j All point clouds of ′ are added to the set S2 = {P l ,…,P n}, repeat the above process until all point clouds in set P are divided, and obtain sets S1, S2…S c .

[0068] Among them, the value ranges of k, l, n and c are all 1, 2…N, and l>k, n>l;

[0069] Among them, in the point cloud P l The method for selecting the reference point is: l After Delaunay triangulation, the average Euclidean distance d is obtained. l Select the initial reference point A j The Euclidean distance of The feature point A j ' as the reference point, wherein the initial reference point A j The reference points are all points outside the plane that are approximately parallel to the laser projection direction when the data acquisition system is scanning.

[0070] S322, calculate the point cloud P in the set S1 k Except for the initial reference point A j In addition, A j is the local feature descriptor value η of each point of the initial reference point; k The point with the largest local feature descriptor value (not unique) is taken as the boundary point to be selected, and the point with the smallest local feature descriptor value (not unique) is taken as the smoothing point to be selected;

[0071] By the adjacent point cloud (point cloud P k and point cloud P k+1 ) to obtain the point cloud P k+1 The boundary point set Φ in k+1 and smooth point set Ω k+1 ; From point cloud With point cloud P k+1 The registration mapping relationship between them is used to obtain the point cloud The boundary point set in and smooth point set

[0072] Among them, the calculation point cloud P k Except for the initial reference point, each point is A j The calculation formula for the local feature descriptor value η of the initial reference point is as follows:

[0073]

[0074] Among them, A j ,A i ∈S,j≠i, M is the point cloud P k The total number of inliers.

[0075] S323, if the number of boundary points to be selected or the number of smooth points to be selected is zero or M, re-execute S321-S322, that is, re-select the initial reference point and recalculate; otherwise, execute S33.

[0076] S33, calculate the boundary point set φ k+1 and smooth point set Ω k+1 The pose transformation matrix corresponding to the midpoint The specific steps are as follows:

[0077] S331, define the pose transformation matrix as At the current timestamp t k+2 Next, let [t k+1 ,t k+2 ] The pose transformation matrix of the data acquisition system during the time period is: Among them, t x ,t y ,t z Respectively represent the translation transformation of the x, y, and z axes in the L coordinate system, θ x ,θ y ,θ z They represent the rotation transformation of the x, y, and z axes in the L coordinate system respectively.

[0078] S332, by combining formula (2) and formula (3), the pose transformation matrix can be solved as follows:

[0079]

[0080] in, is the point cloud P k+1 The boundary point set Φ k+1 Or smooth point set Ω k+1 The pose transformation matrix of any point q in t q Indicates the timestamp (collection time) of point q, is the point cloud P in the L coordinate system kThe coordinates of any point q in the function σ(*,n,m) are the values ​​of the nth to mth items of * ( Representation Matrix The vector consisting of the solutions of the nth to mth terms of the corresponding homogeneous linear equations), n = 1, m = 3, the Rodrigues formula of R is as follows:

[0081] R=e ωθ =I+ωsinθ+ω 2 (1-cosθ)

[0082]

[0083] Among them, the function Ψ(*) is a skew-symmetric matrix taking the variable *, and I is the unit matrix.

[0084] S34, the boundary point set Φ in S33 k+1 and smooth point set Ω k+1 Replace with boundary point set and smooth point set Execute step S33 to calculate the corresponding posture transformation matrix The pose transformation matrix and the pose transformation matrix Converge to get the final pose transformation matrix.

[0085] S35, iterative convergence solution: traverse the set S2…S c , execute S322-S34 to obtain the final posture transformation matrix corresponding to each set, and then obtain the spatial motion estimation curve of the data acquisition system corresponding to the set P, where the horizontal coordinate of the spatial motion estimation curve is the acquisition timestamp, and the vertical coordinate is the final posture transformation matrix in S34.

[0086] S36, repeat S1-S35 until the point cloud P obtained by the last scan of the data acquisition system is obtained. final Corresponding spatial motion estimation curve of the data acquisition system.

[0087] S4, mapping track map:

[0088] S41, unification of the world coordinate system W: In the spatial motion estimation curve obtained in step S36, each time stamp includes corresponding point cloud data containing posture information, and all point cloud data are registered in the world coordinate system W using the real space mapping relationship.

[0089] S42, orbital prior mapping:

[0090] According to the prior feature information in the track application scenario, the point cloud fine registration algorithm is used to fuse the point clouds of adjacent time stamps in the point cloud under the world coordinate system W to realize track mapping, such as Figure 3 shown.

[0091] Among them, the prior feature information includes spatial seamless continuity and local consistency of the rail contour. Within a certain period of time, it can be considered that the weight of logical information (physically) is higher than the weight of measurement information (obtained by the point cloud fine registration algorithm).

[0092] The updating frequency of step S41 should generally be 10 times higher than the updating frequency of step S42, and the field of view overlap during the updating process should be higher than 70%.

[0093] It should be noted that although the operations of the method of the present disclosure are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in this particular order, or that all the operations shown must be performed to achieve the desired results. On the contrary, the steps depicted in the flow chart can be performed in a different order. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps.

Claims

1. A difference mapping and orbit mapping method with weak inertial navigation dependence, characterized in that The following steps are involved: S1, data collection: Use the data collection system to collect and obtain point cloud data around the track, and record the point cloud data obtained by the k-th scan of the data collection system as point cloud P k , and add the point cloud data obtained by each scan to the set, then after N scans, the set P = {P1, P2, ..., P k-1 ,P k ,P k+1 ,…P N }, k = 1, 2…N; S2, define the system coordinate system L and the world coordinate system W: System coordinate system L: The geometric center position of the laser radar in the data acquisition system is the origin position of the system coordinate system L. The positive direction of the z-axis of the system coordinate system L is the same as the direction of the initial posture of the data acquisition system. The y-axis is perpendicular to the z-axis and the positive direction is away from the center of the earth. The x-axis, y-axis and z-axis form a right-handed coordinate system. The system coordinate system constructed by the kth scan is denoted as L k ; The point cloud P k Any point A in i In the system coordinate system L k The coordinates below are expressed as i=1,2,…,M, where M is the point cloud P k The total number of inliers; World coordinate system W: The world coordinate system W completely coincides with the system coordinate system L1 constructed during the first scan of the data acquisition system, where the point cloud P k Any point A in i In the world coordinate system W k The coordinates below are expressed as S3, track point cloud mileage estimation, includes the following steps: S31, point cloud reprojection: Set P = {P1, P2, ..., P k-1 ,P k ,P k+1 ,…P N Each point cloud in} is reprojected to its corresponding scanning end time to obtain the set in: Point Cloud k The scan start time is t k , the scan end time is t k+1 , the point cloud P k Reproject to t k+1 Timestamp, get point cloud Then, the point cloud is obtained from the start time to the end time of the k+1th scan of the data acquisition system and P k+1 ; S32, extract local feature descriptors and obtain point cloud P according to the registration mapping relationship k+1 The boundary point set φ in k+1 and smooth point set Ω k+1 , and point cloud The boundary point set in and smooth point set S321, point cloud P in set P k Select a feature point A j As the initial reference point, the set P contains the initial reference point A j All point clouds are added to the set S1 = {P k ,…,P l }, in the point cloud P l Select a feature point A j ' as the reference point, and the set P contains the reference point A j All point clouds of ′ are added to the set S2 = {P l ,…,P n }; Repeat the above process until all point clouds in set P are divided and sets S1, S2…S are obtained. c ; Among them, the value ranges of k, l, n and c are all 1, 2…N, and l>k, n>l; S322, calculate the point cloud P in the set S1 k Except for the initial reference point A j In addition, A j is the local feature descriptor value η of each point of the initial reference point; k The point with the largest local feature descriptor value is taken as the boundary point to be selected, and the point with the smallest local feature descriptor value is taken as the smoothing point to be selected; From the point cloud P k and point cloud P k+1 The registration mapping relationship between them is used to obtain the point cloud P k+1 The boundary point set φ in k+1 and smooth point set Ω k+1 ; From point cloud With point cloud P k+1 The registration mapping relationship between them is used to obtain the point cloud The boundary point set in and smooth point set Among them, the calculation point cloud P k Except for the initial reference point, each point is A j The calculation formula for the local feature descriptor value η of the initial reference point is as follows: Among them, A j ,A i ∈S,j≠i, M is the point cloud P k The total number of inliers; S323, if the number of boundary points to be selected or the number of smooth points to be selected is zero or M, re-execute S321-S322, that is, re-select the initial reference point and recalculate; otherwise, execute S33; S33, calculate the pose transformation matrix The specific steps are as follows: S331, define the pose transformation matrix as At the current timestamp t k+2 Next, let [t k+1 ,t k+2 ] The pose transformation matrix of the data acquisition system during the time period is: Among them, t x ,t y ,t z Respectively represent the translation transformation of the x, y, and z axes in the L coordinate system, θ x ,θ y ,θ z Respectively represent the rotation transformation of the x, y, and z axes in the L coordinate system; S332, formula (2) and formula (3) are combined to solve the pose transformation matrix: in, is the point cloud P k+1 The boundary point set Φ k+1 Or smooth point set Ω k+1 The pose transformation matrix of any point q in t q Indicates the timestamp of point q, is the point cloud P in the L coordinate system k The coordinates of any point q in the function σ(*,n,m) are the values ​​of the nth to mth items of *. Representation Matrix The corresponding homogeneous linear equation system consists of the solutions of the nth to mth terms), and the Rodrigues formula of R is as follows: R=e ωθ =I+ωsinθ+ω 2 (1-cosθ) Among them, the function Ψ(*) is a skew-symmetric matrix taking the variable *, and I is the identity matrix; S34, the boundary point set Φ in S33 k+1 and smooth point set Ω k+1 Replace with boundary point set and smooth point set Execute step S33 to calculate the corresponding posture transformation matrix The pose transformation matrix and the pose transformation matrix Converge to get the final pose transformation matrix; S35, iterative convergence solution: traverse the set S2…S c , execute S322-S34 to obtain the final posture transformation matrix corresponding to each set, and then obtain the spatial motion estimation curve of the data acquisition system corresponding to the set P, wherein the abscissa of the spatial motion estimation curve is the acquisition timestamp, and the ordinate is the final posture transformation matrix in S34; S36, repeat S1-S35 until the point cloud P obtained by the last scan of the data acquisition system is obtained. final The corresponding spatial motion estimation curve of the data acquisition system; S4, mapping track map: S41, unification of the world coordinate system W: In the spatial motion estimation curve obtained in step S36, each time stamp includes corresponding point cloud data containing posture information, and all point cloud data are registered into the world coordinate system W using the real space mapping relationship; S42, orbital prior mapping: According to the prior feature information in the track application scenario, the point cloud fine registration algorithm is used to fuse the point clouds of adjacent timestamps in the point cloud under the world coordinate system W to realize track mapping.

2. The method for difference mapping and orbit mapping with weak inertial navigation dependence according to claim 1, characterized in that: The data acquisition system in S1 uses a self-propelled RC-SLAM high-precision 3D scanning system, or other 3D scanning systems with dual-axis laser radar.

3. The method for difference mapping and orbit mapping with weak inertial navigation dependence according to claim 1, characterized in that: In S321, in the point cloud P l The method for selecting the reference point is: l After Delaunay triangulation, the average Euclidean distance d is obtained. l Select the initial reference point A j The Euclidean distance The feature point A j ' as the reference point, wherein the initial reference point A j The reference points are all points outside the plane that are approximately parallel to the laser projection direction when the data acquisition system is scanning.

4. The method for difference mapping and orbit mapping with weak inertial navigation dependence according to claim 1, characterized in that: In S42, the priori characteristic information includes spatial seamless continuity and local consistency of rail profile, and within a certain time period, it is considered that the weight of logical information is higher than the weight of measurement information.

5. The method for difference mapping and orbit mapping with weak inertial navigation dependence according to claim 1, characterized in that: The registration frequency of S41 is 10 times that of S42 fusion frequency, and the field of view overlap during the update process is higher than 70%.

6. The method for difference mapping and orbit mapping with weak inertial navigation dependence according to claim 1, characterized in that: In S322, n=1, m=3.

Images