Point cloud map construction and correction method and system based on tunnel mapping

Abstract

The invention discloses a tunnel mapping-based point cloud map construction and correction method and system, and the method comprises the steps: obtaining and preprocessing point cloud data, and merging point cloud key frames according to the corresponding initial poses to obtain a complete initial tunnel point cloud map; obtaining road sign points in the initial tunnel point cloud map; the geometric structure of the track in the tunnel mapping and the installation position of the vehicle-mounted sensor in the train are obtained and utilized, and the pose track of the vehicle-mounted sensor in the tunnel is calculated; and optimizing the initial pose of the point cloud key frame by using the pose track according to the position of the road sign point in the tunnel map to obtain a corrected tunnel point cloud map. According to the method, the precision and the reliability of point cloud map construction can be improved without being limited and influenced by a field environment.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and system for constructing and correcting a point cloud map based on a tunnel construction map. Background Art

[0002] With the acceleration of urbanization, the development and utilization of underground space has gradually become an important part of urban construction. Tunnels, as an effective underground transportation channel, are widely used in subways, light rails and other fields. Among them, tunnel lidar point cloud maps play an important role in tunnel construction and tunnel train automatic driving positioning. The construction and correction methods of tunnel point cloud maps will directly affect the accuracy and reliability of the map, and will also affect the positioning of automatic driving trains and obstacle detection. The existing related point cloud map construction and correction methods are difficult to handle the accumulated errors of the odometer in tunnel scenes where absolute observations are lacking. Therefore, an efficient and accurate point cloud map construction and correction method is needed to improve the accuracy and reliability of point cloud map construction. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for constructing and correcting a point cloud map based on a tunnel construction map, so as to improve the accuracy and reliability of point cloud map construction without being restricted or affected by the on-site environment.

[0004] In order to achieve the above objectives, the present invention is implemented by the following technical solutions:

[0005] A method for constructing and correcting a point cloud map based on a tunnel construction map, comprising:

[0006] Acquire and preprocess point cloud data, merge point cloud keyframes according to their corresponding initial poses to obtain a complete initial tunnel point cloud map;

[0007] Acquire landmark points in the initial tunnel point cloud map;

[0008] Obtain and use the geometric structure of the track in the tunnel construction diagram and the location of the on-board sensor installed in the train to calculate the position trajectory of the on-board sensor in the tunnel;

[0009] By using the posture trajectory and according to the position of the landmark point in the tunnel construction map, the initial posture of the point cloud key frame is optimized to obtain a corrected tunnel point cloud map.

[0010] Optionally, the acquiring and preprocessing the point cloud data includes: using the vehicle-mounted sensor to collect the point cloud data in the tunnel, and filtering and dedistorting the point cloud data.

[0011] Optionally, the vehicle-mounted sensor includes a laser radar.

[0012] Optionally, the initial pose corresponding to the point cloud keyframe is calculated using a laser odometry, and all point cloud keyframes are merged through a point cloud library tool to obtain a complete initial tunnel point cloud map.

[0013] Optionally, a method of extracting prominent three-dimensional features of the tunnel surface is used to find landmark points in the initial tunnel point cloud map, and the landmark points include kilometer landmark points and beacon points.

[0014] Optionally, the tunnel construction map is used to construct a three-dimensional trajectory of the track in the tunnel, and then the position trajectory of the on-board sensor moving in the tunnel is calculated and converted according to the position of the on-board sensor installed in the train.

[0015] Optionally, the pose trajectory of the vehicle-mounted sensor in the tunnel and the initial pose trajectory of the point cloud key frame are used to construct a target equation to optimize the initial pose of the point cloud key frame according to the corresponding landmark point positions in the tunnel construction map, and finally the point cloud key frames are merged to obtain a corrected tunnel point cloud map.

[0016] Optionally, the corresponding landmark point position in the tunnel construction map is to find the absolute position of the landmark point relative to the vehicle-mounted sensor in the tunnel construction map.

[0017] Optionally, the optimized point cloud key frames are merged using a laser odometry to obtain a revised point cloud map of the tunnel.

[0018] Optionally, the target equation is a pose target equation for constructing landmark points and point cloud key frames using a pose graph optimization method, and the expression of the target equation is as follows:

[0019]

[0020] Among them, the residual It can be defined as:

[0021]

[0022] In the formula, represents the pose matrix of the nth landmark point in the tunnel at the corresponding position of the lidar trajectory, T n represents the key frame pose matrix that can observe the nth landmark point mentioned above. The superscript N represents the total number of N landmark points. ln(·) ∨ The function represents the logarithmic mapping of the pose matrix form T to the corresponding Lie algebraic form, which corresponds to a six-dimensional vector; The information matrix representing the constraint factors of the tunnel landmark points can be adjusted according to the accuracy of the tunnel construction map. It can usually be set as a diagonal matrix, and the diagonal elements are used to represent the weights of the corresponding residuals; It is expressed as:

[0023]

[0024] Among them, t x ,t y ,t z Corresponding to the first three-dimensional translation components, φ r ,φ p ,φ y Corresponding to the rotation component.

[0025] On the other hand, the present invention also provides a point cloud map construction and correction system based on a tunnel construction map, including: an initial point cloud map construction module, used to acquire and preprocess point cloud data, and merge point cloud key frames according to corresponding initial poses to obtain a complete initial tunnel point cloud map; a landmark point extraction module, used to acquire landmark points in the initial tunnel point cloud map; a tunnel construction map track modeling module, used to acquire and utilize the geometric structure of the track in the tunnel construction map and the position of the on-board sensor installed in the train, to calculate the pose trajectory of the on-board sensor in the tunnel; a point cloud map correction module, used to utilize the pose trajectory and optimize the initial pose of the point cloud key frame according to the position of the landmark point in the tunnel construction map to obtain a corrected tunnel point cloud map.

[0026] On the other hand, the present invention further provides an electronic device, including a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method described above is implemented.

[0027] In other aspects, the present invention further provides a readable storage medium, wherein the readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0028] Compared with the prior art, the present invention has at least the following technical effects:

[0029] The present invention uses the tunnel construction map to correspond the tunnel landmark points with obvious three-dimensional structures in the tunnel with the precise coordinates of the landmark points in the tunnel construction map, introduces global information to optimize the tunnel point cloud map, and eliminates the accumulated errors in the point cloud map construction process; thus, the accuracy and reliability of point cloud map construction are improved without being restricted or affected by the on-site environment.

[0030] The present invention takes into account the difference between the direction of the track trajectory and the direction of the train laser radar, and converts the train track trajectory in the tunnel construction map according to the geometric relationship to obtain the motion trajectory of the on-board laser radar, which is used for point cloud map correction to improve the accuracy of point cloud map construction and correction.

[0031] The present invention does not introduce other additional sensor information, and can reduce the consumption of real-time computing resources.

[0032] The present invention does not require other auxiliary equipment, and can reduce equipment cost and installation cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A schematic diagram of a process of constructing and correcting a point cloud map based on a tunnel construction map provided by an embodiment of the present invention;

[0034] Figure 2 A structural block diagram of a point cloud map construction and correction system based on a tunnel construction map provided by an embodiment of the present invention;

[0035] Figure 3 A schematic diagram of initial point cloud map construction and landmark point extraction provided by an embodiment of the present invention;

[0036] Figure 4 A schematic diagram of a tunnel construction diagram track center trajectory to a train laser radar trajectory provided by an embodiment of the present invention;

[0037] Figure 5 A schematic diagram of key frame pose optimization corresponding to the landmark points in the tunnel construction map and the initial point cloud map provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following is a further detailed description of a method and system for constructing and correcting a point cloud map based on a tunnel construction map proposed by the present invention in combination with the accompanying drawings and specific embodiments. According to the following description, the advantages and features of the present invention will be clearer. It should be noted that the accompanying drawings are in a very simplified form and use non-precise proportions, which are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In order to make the purposes, features and advantages of the present invention more obvious and easy to understand, please refer to the accompanying drawings. It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention, so they have no technical substantive significance. Any structural modification, change in proportional relationship or adjustment of size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope of the technical content disclosed by the present invention.

[0039] As described in the background technology, the current methods for constructing and correcting point cloud maps in tunnels mainly include the following:

[0040] One solution is to introduce information from IMU (Inertial Measurement Unit) and odometer. IMU can provide information such as angular velocity and acceleration, while odometer can provide displacement information of the train. By fusing IMU and odometer information with lidar data, the accuracy of point cloud map construction in tunnels can be improved.

[0041] Another solution is to introduce auxiliary equipment such as UWB (Ultra Wide Band) and reflective columns. Auxiliary means can provide additional information to assist in the construction of point cloud maps and improve the accuracy of point cloud mapping.

[0042] However, the above method has some problems in tunnel scenarios, mainly including the following points:

[0043] First, in tunnel scenarios, there is a lack of absolute observation. Although the introduction of IMU and odometer information can improve the accuracy of point cloud map construction, its accuracy and stability are still limited by sensor noise and errors. In addition, the cumulative error of IMU and odometer will lead to cumulative errors in the point cloud map construction process, which will affect the subsequent point cloud positioning and obstacle detection effects.

[0044] In the absence of global information in the tunnel, auxiliary means such as reflective columns can only enrich the point cloud features of the tunnel scene and cannot eliminate the accumulated errors in the mapping process. In addition, auxiliary means such as UWB or reflective columns will greatly increase the external deployment cost, and the implementation conditions are easily restricted and affected by the on-site environment.

[0045] In summary, the existing methods for constructing and correcting tunnel point cloud maps for rail transit have problems such as difficulty in eliminating cumulative errors, low accuracy, high cost of additional equipment, and poor versatility.

[0046] In view of this, if Figure 1 As shown, this embodiment provides a method for constructing and correcting a point cloud map based on a tunnel construction map, including: acquiring and preprocessing point cloud data, and merging point cloud key frames according to their corresponding initial poses to obtain a complete initial tunnel point cloud map.

[0047] In this embodiment, each frame of point cloud data periodically extracted at preset intervals from the point cloud data is used as the point cloud key frame.

[0048] In this embodiment, the point cloud key frames are calculated using a point cloud registration method to obtain the initial poses of all point cloud key frames.

[0049] Obtaining landmark points in the initial tunnel point cloud map.

[0050] The geometric structure of the track in the tunnel construction diagram and the position of the on-board sensors installed in the train are obtained and used to calculate the position trajectory of the on-board sensors in the tunnel.

[0051] By using the posture trajectory and according to the position of the landmark point in the tunnel construction map, the initial posture of the point cloud key frame is optimized to obtain a corrected tunnel point cloud map.

[0052] This embodiment uses the tunnel construction map to correspond the tunnel landmark points with obvious three-dimensional structures in the tunnel with the precise coordinates of the landmark points in the tunnel construction map, introduces global information to optimize the tunnel point cloud map, and eliminates the accumulated errors in the point cloud map construction process; it achieves the improvement of the accuracy and reliability of point cloud map construction without being restricted or affected by the on-site environment.

[0053] Please continue to refer to Figure 1 As shown, in this embodiment, the acquiring and preprocessing of point cloud data includes: using the vehicle-mounted sensor to collect the point cloud data in the tunnel, and filtering and dedistorting the point cloud data.

[0054] Please continue to refer to Figure 1 As shown, in this embodiment, the vehicle-mounted sensor includes a laser radar. This embodiment does not introduce other additional sensor information, which can reduce the consumption of real-time computing resources.

[0055] Please continue to refer to Figure 1 As shown, in this embodiment, the laser odometry is used to calculate the initial pose corresponding to the point cloud keyframe, and all the point cloud keyframes are merged through the point cloud library tool (PCL) to obtain a complete initial tunnel point cloud map. It can be understood that the laser odometry is a commonly used method in laser SLAM (Simultaneous Localization and Mapping), which is generally based on the iterative closest point algorithm (ICP) or the normal distribution transform (NDT) to perform point cloud registration between two frames of point clouds to obtain the transformation relationship between the two point clouds.

[0056] Please continue to refer to Figure 1 As shown, in this embodiment, the tunnel surface protruding three-dimensional feature extraction method is adopted to find the landmark points in the initial tunnel point cloud map, and the landmark points include kilometer landmark points and beacon points.

[0057] Please continue to refer to Figure 1 As shown, in this embodiment, the tunnel construction map is used to construct the three-dimensional trajectory of the track in the tunnel, and then according to the position of the on-board sensor installed in the train, the posture trajectory of the on-board sensor moving in the tunnel is calculated and converted.

[0058] Please continue to refer to Figure 1As shown, in this embodiment, the posture trajectory of the vehicle-mounted sensor in the tunnel and the initial posture trajectory of the point cloud key frame are utilized, and according to the corresponding landmark point positions in the tunnel construction map, a target equation is constructed to optimize the initial posture of the point cloud key frame, and finally the point cloud key frames are merged to obtain a corrected tunnel point cloud map.

[0059] Please continue to refer to Figure 1 As shown, in this embodiment, the corresponding landmark point position in the tunnel construction map is to find the absolute position of the landmark point relative to the vehicle-mounted sensor in the tunnel construction map.

[0060] Please continue to refer to Figure 1 As shown, in this embodiment, the optimized point cloud key frames are merged using a laser odometer to obtain the corrected tunnel point cloud map.

[0061] Please continue to refer to Figure 1 As shown, in this embodiment, the target equation is a pose target equation for constructing landmark points and point cloud key frames using a pose graph optimization method:

[0062]

[0063] Among them, the residual It can be defined as:

[0064]

[0065] In the formula, represents the pose matrix of the nth landmark point in the tunnel at the corresponding position of the lidar trajectory, T n represents the key frame pose matrix that can observe the nth landmark point mentioned above. The superscript N represents the total number of N landmark points. ln(·) ∨ The function represents the logarithmic mapping of the pose matrix form T to the corresponding Lie algebraic form, which corresponds to a six-dimensional vector; The information matrix representing the constraint factors of tunnel landmark points can be adjusted according to the accuracy of the tunnel construction map. It can usually be set as a diagonal matrix, and the diagonal elements are used to represent the weights of the corresponding residuals. It is expressed as:

[0066]

[0067] Among them, t x ,t y ,t z Corresponding to the first three-dimensional translation components, φ r ,φ p ,φ y corresponds to the rotation component.

[0068] This embodiment takes into account the difference between the direction of the track trajectory and the direction of the laser radar, and converts the train track trajectory in the tunnel construction map according to the geometric relationship to obtain the motion trajectory of the on-board laser radar, which is used for point cloud map correction to improve the accuracy of point cloud map construction and correction.

[0069] like Figure 2 As shown, on the other hand, the present invention also provides a point cloud map construction and correction system based on a tunnel construction map, comprising:

[0070] The initial point cloud map construction module S01 is used to obtain and pre-process the point cloud data, merge the point cloud key frames according to the corresponding initial poses to obtain a complete initial tunnel point cloud map;

[0071] A landmark point extraction module S02 is used to obtain landmark points in the initial tunnel point cloud map;

[0072] The tunnel construction map track modeling module S03 is used to obtain and use the geometric structure of the track in the tunnel construction map and the position of the on-board sensor installed in the train to calculate the position trajectory of the on-board sensor in the tunnel;

[0073] The point cloud map correction module S04 is used to optimize the initial pose of the point cloud key frame by using the pose trajectory and according to the position of the landmark point in the tunnel construction map to obtain a corrected tunnel point cloud map.

[0074] This embodiment does not require other auxiliary equipment, which can reduce equipment cost and installation cost.

[0075] In order to better understand the above embodiment, a more specific embodiment is described below:

[0076] like Figure 3 As shown in the figure, the original tunnel point cloud data is collected by the laser radar installed in front of the train. The distance between adjacent platforms is about 1000m, and a frame of original point cloud data is extracted at intervals of about 3m as a key frame. Figure 3 The content shown is several consecutive frames of key frames and feature points in a certain adjacent station, and the above key frames are pre-processed by filtering, removing invalid points, de-distorting, extracting features, etc.

[0077] The laser odometry is used to obtain the initial poses of all point cloud key frames using a point cloud registration method based on the iterative closest point method.

[0078] The calculation formula is:

[0079]

[0080] Among them, P s represents the source point cloud, P t represents the target point cloud; R *,t * Represents the optimal estimated pose result.

[0081] The initial poses of all point cloud key frames {L0,,,L n}, where L n =(R n ,t n ),use Figure 3 The triangle in the figure is represented by n Represents the rotation matrix of the nth frame point cloud relative to the initial point cloud coordinate system; t n Represents the translation vector of the nth frame point cloud relative to the initial point cloud coordinate system.

[0082] According to the initial poses of all point cloud key frames, about three hundred point cloud key frames are spliced ​​and merged according to their corresponding initial poses, so as to obtain an initial tunnel point cloud map with a complete tunnel feature structure.

[0083] It can be understood that the point cloud key frame data here is taken from the original frame point cloud data. Since the amount of original frame point cloud data of the lidar is too large, 10 frames of point cloud can be received per second, and each frame of point cloud has 2w+ (i.e. 20,000+) points. Therefore, a frame of point cloud data is selected as the point cloud key frame at a fixed distance to ensure that the device calculation time is short and the map point cloud is evenly distributed.

[0084] Then, continue to search for kilometer markers and beacon points (positions of road signs) with prominent geometric features in the tunnel in the initial tunnel point cloud map. n , Figure 3 As shown in the five-pointed star.

[0085] It can be understood that the kilometer marker is located on the side of the tunnel wall, and the beacon point is in the middle of the track. Only the kilometer marker and the beacon have clear location marking points in the tunnel construction map and can be directly observed and identified in the initial tunnel point cloud map.

[0086] And continue to find the key frame K corresponding to the above features in the initial tunnel point cloud map n (i.e., the key frame point cloud of beacons or kilometer markers can be observed).

[0087] It can be understood that the features here are the features of beacons and kilometer markers that can be directly observed in the initial tunnel point cloud map. For example, the beacon is located in the middle of the railroad sleepers and protrudes, which can be directly observed in the initial tunnel point cloud map. The kilometer marker is a square metal plate on the side of the tunnel wall with a kilometer marker number. It is a highly reflective target in the initial tunnel point cloud map and can be observed directly. The specific kilometer number can be seen in the corresponding video, so that the exact location can be found in the tunnel construction map.

[0088] Furthermore, assuming that the distance between two adjacent stations is 1 km, there are about 300 point cloud keyframes. Beacons and kilometer landmarks are relatively sparse, with about ten kilometer landmarks and about 20 beacons. Not all point cloud keyframes can observe beacons or kilometer landmarks. We need to find the keyframe that can observe one of the landmarks (i.e., keyframe K n ), the pose of such key frames can be used as optimization variables in the subsequent tunnel point cloud map optimization process.

[0089] After that, use a tool that can view the 3D point cloud (such as the pcl_viewer tool, which can select a beacon point or a kilometer mark point (road point) in the point cloud and obtain its coordinate value in the keyframe point cloud coordinate system) to view the point cloud keyframe K n Middle kilometer mark and beacon point F n The transformation matrix T to the origin of the point cloud coordinates of this frame Fn->Kn Used for the conversion of the laser radar center coordinate trajectory in the subsequent tunnel construction map.

[0090] like Figure 4 As shown in the figure, the track path diagram between adjacent stations can be calculated based on the side view and top view in the tunnel construction diagram, which is a three-dimensional line track in three-dimensional space. Since the laser radar is located at the head of the train, the movement direction of the laser radar does not coincide with the direction of the track in the tunnel. It is necessary to calculate the laser radar according to its installation position at the head of the train. Figure 4 The geometric relationship in the figure is used to infer the new motion trajectory of the lidar in the tunnel when the train moves forward.

[0091] It is understandable that the geometric relationship here can be expressed by Figure 3 In the figure, the position relationship between the laser radar installation position and the center of the train chassis is obtained, and the center track of the track can be obtained through the tunnel construction map. Figure 4 The schematic diagram of the position relationship in the figure can be used to obtain the actual motion trajectory of the laser radar.

[0092] Calculation formula:

[0093]

[0094] in, is the position point on the motion trajectory of the front laser radar, represents the track center position point in the tunnel construction diagram, T Tunnel-＞LiDAR Represents the positional relationship between the laser radar installation position and the center of the locomotive chassis.

[0095] like Figure 5 As shown in Figure 1, based on the precise marking positions of the landmarks in the tunnels between stations in the tunnel construction diagram, the precise position of the landmarks in the track path can be calculated. nT , and then through Figure 3 The transformation matrix T calculated in Fn->Kn , the precise position of the landmark point in the track path F n T Reconvert to the actual position K of the corresponding laser radar in the orbit path n Tunnel .

[0096] The specific calculation formula is:

[0097] K n T =F n T ·(T Fn-＞Kn ) -1 (1.12)

[0098] According to the key frame pose K of the landmark point that can be observed in the above two processes n and the actual pose K n Tunnel Construct the target equation:

[0099]

[0100] Among them, the residual It can be defined as:

[0101]

[0102] In the formula, represents the pose matrix of the nth landmark point in the tunnel at the corresponding position of the lidar trajectory, T n represents the key frame pose matrix that can observe the nth landmark point mentioned above. The superscript N represents the total number of N landmark points. ln(·) ∨ The function represents the logarithmic mapping of the pose matrix form T to the corresponding Lie algebraic form, which corresponds to a six-dimensional vector; The information matrix representing the constraint factors of tunnel landmark points can be adjusted according to the accuracy of the tunnel construction map. It can usually be set as a diagonal matrix, and the diagonal elements are used to represent the weights of the corresponding residuals.

[0103] Solve the optimization target equation (for example, use Ceres Solver (a nonlinear optimization library developed by Google) to solve the optimization target equation), obtain the key frame pose of the optimized landmark points, and re-merge them to obtain the corrected point cloud map.

[0104] In other aspects, the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the method described above is implemented.

[0105] In other aspects, the present invention further provides a readable storage medium, wherein the readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0106] In summary, the method provided in this embodiment includes: first, using sensors such as laser radar to collect point cloud data in the tunnel, performing preprocessing operations such as filtering and dedistortion, and then using a laser odometer to merge the point cloud key frames according to their corresponding initial poses to obtain a complete initial tunnel point cloud map. And find the corresponding kilometer mark points and beacon points in the initial tunnel point cloud map. At the same time, the tunnel construction map is used to construct the three-dimensional trajectory of the track in the tunnel, and then according to the position of the laser radar installed in the train, the posture trajectory of the laser radar moving in the tunnel is calculated and converted. Using the posture trajectory of the laser radar in the tunnel obtained by the above calculation and the initial posture trajectory of the key frame, according to the corresponding landmark point positions in the tunnel construction map, the target equation is constructed to optimize the initial posture of the key frame mentioned above, and finally the key frame point cloud is merged to obtain the corrected point cloud map, so as to eliminate the accumulated error of the laser odometer and improve the accuracy and reliability of the point cloud map.

[0107] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0108] It should be noted that the devices and methods disclosed in the embodiments of this article can also be implemented in other ways. The device implementation described above is only schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of this article. In this regard, each box in the flowchart or block diagram can represent a part of a module, program or code, and the module, program segment or part of the code contains one or more executable instructions for implementing the specified logical function, and the module, program segment or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented by a dedicated hardware-based system for performing a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0109] In addition, the functional modules in the various embodiments of this document may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0110] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be appreciated that the above description should not be considered as a limitation of the present invention. After reading the above content, it will be apparent to those skilled in the art that various modifications and substitutions of the present invention will occur. Therefore, the protection scope of the present invention should be limited by the appended claims.

Claims

1. A method for constructing and correcting a point cloud map based on a tunnel construction map, characterized in that: include: Acquire and preprocess point cloud data, merge point cloud keyframes according to their corresponding initial poses to obtain a complete initial tunnel point cloud map; Acquire landmark points in the initial tunnel point cloud map; Obtain and use the geometric structure of the track in the tunnel construction diagram and the location of the on-board sensor installed in the train to calculate the position trajectory of the on-board sensor in the tunnel; By using the posture trajectory and according to the position of the landmark point in the tunnel construction map, the initial posture of the point cloud key frame is optimized to obtain a corrected tunnel point cloud map.

2. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 1, characterized in that: The acquiring and preprocessing of point cloud data includes: using the vehicle-mounted sensor to collect the point cloud data in the tunnel, and filtering and de-distorting the point cloud data.

3. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 1, characterized in that: The vehicle-mounted sensor includes a laser radar.

4. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 3, characterized in that: The laser odometry is used to calculate the initial pose corresponding to the point cloud keyframe, and all point cloud keyframes are merged through the point cloud library tool to obtain a complete initial tunnel point cloud map.

5. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 1, characterized in that: The tunnel surface protruding three-dimensional feature extraction method is adopted to find the landmark points in the initial tunnel point cloud map, and the landmark points include kilometer landmark points and beacon points.

6. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 1, characterized in that: The tunnel construction map is used to construct a three-dimensional trajectory of the track in the tunnel, and then the position trajectory of the on-board sensor moving in the tunnel is calculated and converted according to the position where the on-board sensor is installed in the train.

7. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 1, characterized in that: By using the pose trajectory of the vehicle-mounted sensor in the tunnel and the initial pose trajectory of the point cloud key frame, according to the corresponding landmark point positions in the tunnel construction map, the target equation is constructed to optimize the initial pose of the point cloud key frame, and finally the point cloud key frame is merged to obtain the corrected tunnel point cloud map.

8. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 7, characterized in that: The corresponding landmark position in the tunnel construction map is to find the absolute position of the landmark relative to the vehicle-mounted sensor in the tunnel construction map.

9. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 7, characterized in that: The optimized point cloud key frames are merged using a laser odometry to obtain a revised point cloud map of the tunnel.

10. The method for constructing and correcting a point cloud map based on a tunnel construction map according to claim 7, characterized in that: The target equation is a pose target equation for constructing landmark points and point cloud key frames using a pose graph optimization method. The expression of the target equation is as follows: Among them, the residual It can be defined as: In the formula, represents the pose matrix of the nth landmark point in the tunnel at the corresponding position of the lidar trajectory, T n represents the key frame pose matrix that can observe the nth landmark point mentioned above. The superscript N represents the total number of N landmark points. ln(·) ∨ The function represents the logarithmic mapping of the pose matrix form T to the corresponding Lie algebraic form, which corresponds to a six-dimensional vector; The information matrix representing the constraint factors of the tunnel landmark points can be adjusted according to the accuracy of the tunnel construction map. It can usually be set as a diagonal matrix, and the diagonal elements are used to represent the weights of the corresponding residuals; It is expressed as: Among them, t x ,t y ,t z Corresponding to the first three-dimensional translation components, φ r ,φ p ,φ y Corresponding to the rotation component.

11. A point cloud map construction and correction system based on tunnel construction map, characterized in that: include: The initial point cloud map construction module is used to obtain and preprocess the point cloud data, merge the point cloud key frames according to the corresponding initial poses to obtain a complete initial tunnel point cloud map; A landmark point extraction module, used to obtain landmark points in the initial tunnel point cloud map; The tunnel construction map track modeling module is used to obtain and use the geometric structure of the track in the tunnel construction map and the position of the on-board sensor installed in the train to calculate the position trajectory of the on-board sensor in the tunnel; The point cloud map correction module is used to optimize the initial pose of the point cloud key frame by using the pose trajectory and according to the position of the landmark point in the tunnel construction map to obtain a corrected tunnel point cloud map.

12. An electronic device, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method according to any one of claims 1 to 10 is implemented.

13. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images