Vector map updating method, system and device in unmanned mine scene and medium

Abstract

The invention relates to the technical field of unmanned driving, in particular to a vector map updating method, system and device in an unmanned mine scene and a medium. The method comprises the following steps: firstly, preprocessing original point cloud data according to the current pose of the unmanned vehicle; filtering a dynamic object, and calculating a terrain contour; and finally, performing Boolean operation on the terrain contour and the high-precision map to obtain incremental update of the high-precision map, thereby realizing light-weight and high-efficiency update of the high-precision map. The problem that the updating efficiency is low when a vehicle contour scheme is adopted is solved. The updating range is only limited to the measurement distance of the laser radar, and a large-range map can be updated through one-time calculation; the calculation is simple and the calculation efficiency is high.

Description

Technical Field

[0001] The present invention relates to the field of unmanned driving technology, and in particular to a vector map updating method, system, device and medium in an unmanned mine scenario. Background Art

[0002] High-precision maps accurately represent the road ahead and the surrounding environment, a key element to the success of autonomous driving. Common HD map formats include Lanelet2 and OpenDrive. Mining environments are typically unstructured environments. As mining and soil disposal operations progress, the terrain constantly changes, posing significant challenges to environmental mapping. Mining and soil disposal areas are particularly prone to constant topographical changes. In other areas, such as operational roads, topographical changes are relatively minor. Therefore, there is an urgent need for HD map update methods tailored to mining and soil disposal areas.

[0003] The patent application number is "202311783826.5", and the name is "A method and system for updating the boundary of unmanned driving maps in open-pit mines". It uses Boolean operations between vehicle contours and existing maps to achieve map updates; only vehicle contour information is used to update the map. Due to the limited coverage area of the vehicle contour, the update efficiency is low, and it is easy to cause map fragmentation, holes, and other phenomena. Summary of the Invention

[0004] In response to the problem of low update efficiency in existing methods of map updating using vehicle contours, the present invention proposes a vector map update method, system, equipment and medium in an unmanned mining scenario. The method first pre-processes the original point cloud data according to the current position of the unmanned vehicle; then filters dynamic objects and calculates the terrain contour; finally, Boolean operations are performed on the terrain contour and the high-precision map to obtain an incremental update of the high-precision map, thereby realizing a lightweight and efficient update of the high-precision map.

[0005] The specific implementation contents of the present invention are as follows:

[0006] A vector map update method for unmanned mining scenarios first pre-processes raw point cloud data based on the current position of the unmanned vehicle. Then, dynamic objects are filtered out and the terrain contour is calculated based on the pre-processed raw point cloud data. Finally, a Boolean operation is performed on the terrain contour and the high-precision map to obtain an incremental update of the high-precision map.

[0007] In order to better implement the present invention, further, the vector map updating method in the unmanned mine scenario specifically includes the following steps:

[0008] Step S1: pre-processing the original point cloud data according to the current position of the unmanned vehicle obtained from the unmanned positioning system;

[0009] Step S2: creating a point cloud coordinate array based on the pre-processed original point cloud data, connecting the points recorded in the point cloud coordinate array to obtain a contour polygon, and using it as the terrain contour;

[0010] Step S3: Obtain the area closest to the current vehicle position from the high-precision map, describe the area boundary as an inaccessible polygon, perform Boolean operation on the inaccessible polygon and the outline polygon, and obtain an incremental update of the high-precision map.

[0011] In order to better implement the present invention, further, step S1 specifically includes the following steps:

[0012] Step S11: obtaining the current position of the unmanned vehicle from the unmanned positioning system;

[0013] Step S12: transforming the acquired frame of original point cloud data into a world coordinate system according to the current posture of the unmanned vehicle;

[0014] Step S13: Continuously accumulating multiple frames of point cloud according to a set time period to obtain merged point cloud data;

[0015] Step S14: performing voxel statistical filtering and voxel downsampling on the merged point cloud data to obtain filtered merged point cloud data;

[0016] Step S15: performing Euclidean clustering segmentation on the filtered merged point cloud data to obtain merged point cloud data after cluster segmentation;

[0017] Step S16: removing the ground data from the merged point cloud data after clustering and segmentation according to the normal vector to obtain the pre-processed original point cloud data.

[0018] In order to better implement the present invention, further, step S2 specifically includes the following steps:

[0019] Step S21: creating a coordinate point array based on the pre-processed original point cloud data;

[0020] Step S22: traverse the pre-processed original point cloud data and convert the point coordinates into polar coordinates;

[0021] Step S23: Call the round() function to calculate the index id according to the polar coordinates;

[0022] Step S24: updating the contour information at the index id corresponding to the coordinate point array;

[0023] Step S25: Connect the points recorded in the updated coordinate point array in a counterclockwise direction to form a contour polygon, which is used as the terrain contour.

[0024] In order to better implement the present invention, further, step S3 specifically includes the following steps:

[0025] Step S31: searching the high-precision map for the area closest to the current vehicle location, and describing the area boundary as an inaccessible polygon; the area includes a mining area and a soil dumping area;

[0026] Step S32: Perform a Boolean operation on the inaccessible polygon and the outline polygon to obtain an incremental update of the high-precision map.

[0027] In order to better implement the present invention, further, when the area is a mining area, the specific operation of step S32 is: first, the inaccessible polygon and the outline polygon are subjected to geometric Boolean operation to obtain P new =P old -P measure Then use P new Instead of P old Update high-precision maps; among them, P measure is the outline polygon, P old A polygon that is not accessible.

[0028] In order to better implement the present invention, further, when the area is a soil dumping area, the specific operation of step S32 is: starting from the current vehicle position, starting from 0°, every 1° until 360°, make a ray, if the ray is in contact with the inaccessible polygon P old and the contour polygon P measure There are intersection points, then for the inaccessible polygon P old Do the following: Let the ray and P old The intersection point of the ray and the contour polygon P is P1. measure The intersection point is P2, execute: P1 = P1 × 0.8 + 0.2 × P2, and use P1 to replace the inaccessible polygon P old The nearest point in .

[0029] Based on the above-mentioned vector map updating method in an unmanned mine scenario, in order to better implement the present invention, a vector map updating system in an unmanned mine scenario is further proposed, which is used to execute the above-mentioned vector map updating method in an unmanned mine scenario; the system includes a point cloud preprocessing module, a terrain contour calculation module, and a map incremental update module;

[0030] The point cloud preprocessing module is used to preprocess the original point cloud data according to the current posture of the unmanned vehicle;

[0031] The terrain contour calculation module is used to filter dynamic objects and calculate the terrain contour based on the pre-processed original point cloud data;

[0032] The map incremental update module is used to perform Boolean operations on the terrain contour and the high-precision map to obtain an incremental update of the high-precision map.

[0033] Based on the above-mentioned vector map update method in the unmanned mine scenario, in order to better implement the present invention, an electronic device is further proposed, including a memory and a processor; a computer program is stored on the memory; when the computer program is executed on the processor, the above-mentioned vector map update method in the unmanned mine scenario is implemented.

[0034] Based on the above-mentioned vector map update method in the unmanned mine scenario, in order to better implement the present invention, a computer-readable storage medium is further proposed, on which computer instructions are stored; when the computer instructions are executed on the above-mentioned electronic device, the above-mentioned vector map update method in the unmanned mine scenario is implemented.

[0035] The present invention has the following beneficial effects:

[0036] (1) The present invention proposes a lightweight and efficient method for updating high-precision maps of mining areas and soil discharge areas, which solves the problem of low updating efficiency of the vehicle contour scheme.

[0037] (2) The present invention is simple to calculate and has high efficiency. A large-scale map can be updated in one calculation. The updated range is limited to the measurement distance of the laser radar. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a schematic diagram of the point cloud provided by the present invention.

[0039] Figure 2 This is a polar coordinate schematic diagram provided by the present invention.

[0040] Figure 3 This is a schematic diagram of an empty coordinate point array A provided by the present invention.

[0041] Figure 4 This is a schematic diagram of the updated profile information provided by the present invention.

[0042] Figure 5 This is a schematic diagram of the terrain contour provided by the present invention.

[0043] Figure 6 This is a schematic diagram of the updated soil discharge profile provided by the present invention. DETAILED DESCRIPTION

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. It should be understood that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, and therefore should not be regarded as limiting the scope of protection. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technical personnel in this field without making creative work are within the scope of protection of the present invention.

[0045] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "disposed," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections, electrical connections; direct connections, indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0046] Example 1:

[0047] This embodiment proposes a vector map update method for an unmanned mine scenario. First, the raw point cloud data is preprocessed based on the current position of the unmanned vehicle. Then, dynamic objects are filtered out based on the preprocessed raw point cloud data, and the terrain contour is calculated. Finally, a Boolean operation is performed on the terrain contour and the high-precision map to obtain an incremental update of the high-precision map. The method specifically includes the following steps:

[0048] Step S1: pre-processing the original point cloud data according to the current position of the unmanned vehicle obtained from the unmanned positioning system;

[0049] The step S1 specifically includes the following steps:

[0050] Step S11: obtaining the current position of the unmanned vehicle from the unmanned positioning system;

[0051] Step S12: transforming the acquired frame of original point cloud data into a world coordinate system according to the current posture of the unmanned vehicle;

[0052] Step S13: Continuously accumulating multiple frames of point cloud according to a set time period to obtain merged point cloud data;

[0053] Step S14: performing voxel statistical filtering and voxel downsampling on the merged point cloud data to obtain filtered merged point cloud data;

[0054] Step S15: performing Euclidean clustering segmentation on the filtered merged point cloud data to obtain merged point cloud data after cluster segmentation;

[0055] Step S16: removing the ground data from the merged point cloud data after clustering and segmentation according to the normal vector to obtain the pre-processed original point cloud data.

[0056] Step S2: Based on the pre-processed original point cloud data, a coordinate point array is established, and the points recorded in the coordinate point array are connected to obtain a contour polygon, which is used as the terrain contour;

[0057] The step S2 specifically includes the following steps:

[0058] Step S21: creating a coordinate point array based on the pre-processed original point cloud data;

[0059] Step S22: traverse the pre-processed original point cloud data and convert the point coordinates into polar coordinates;

[0060] Step S23: According to the polar coordinates, call the round() function to calculate the index id;

[0061] Step S24: updating the contour information at the index id corresponding to the coordinate point array;

[0062] Step S25: Connect the points recorded in the updated coordinate point array in a counterclockwise direction to form a contour polygon, which is used as the terrain contour.

[0063] Step S3: Obtain the area closest to the current vehicle position from the high-precision map, describe the area boundary as an inaccessible polygon, perform Boolean operation on the inaccessible polygon and the outline polygon, and obtain an incremental update of the high-precision map.

[0064] The step S3 specifically includes the following steps:

[0065] Step S31: searching the high-precision map for the area closest to the current vehicle location, and describing the area boundary as an inaccessible polygon; the area includes a mining area and a soil dumping area;

[0066] Step S32: Perform a Boolean operation on the inaccessible polygon and the outline polygon to obtain an incremental update of the high-precision map.

[0067] When the area is a mining area, the specific operation of step S32 is: first, perform geometric Boolean operation on the inaccessible polygon and the outline polygon to obtain P new =P old -P measure Then use P new Instead of P old Update high-precision maps; among them, P measure is the outline polygon, P old A polygon that is not accessible.

[0068] When the area is a soil dumping area, the specific operation of step S32 is: starting from the current vehicle position, starting from 0°, every 1° until 360°, make a ray, if the ray intersects with the inaccessible polygon P old and the contour polygon P measure There are intersection points, then for the inaccessible polygon P old Do the following: Let the ray and P old The intersection point of the ray and the contour polygon P is P1. measure The intersection point is P2, execute: P1 = P1 × 0.8 + 0.2 × P2, and use P1 to replace the inaccessible polygon P old The nearest point in .

[0069] Working Principle: This embodiment first pre-processes the raw point cloud data based on the current position of the unmanned vehicle; then filters dynamic objects and calculates the terrain contour; finally, Boolean operations are performed on the terrain contour and the high-precision map to obtain an incremental update of the high-precision map, achieving lightweight and efficient updates to the high-precision map.

[0070] Example 2:

[0071] This embodiment is based on the above embodiment 1. Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 、 Figure 6 As shown in the figure, the lanelet2 high-precision map is used as an example for detailed explanation.

[0072] This embodiment mainly uses an on-board laser sensor to scan and perceive the terrain features around the vehicle body, filters out dynamic objects, calculates the terrain contour, and then performs a Boolean operation on this contour and the Lanelet2 high-precision map to obtain an incremental update to the high-precision map. Specifically, the following steps are included:

[0073] Step S1: Point cloud preprocessing.

[0074] Step S11: transforming a frame of original point cloud data into a world coordinate system based on the current posture of the unmanned vehicle, as determined by the unmanned vehicle positioning system;

[0075] Step S12: continuously accumulating multiple frames of point clouds over a period of time and merging them into one point cloud;

[0076] Step S13: performing voxel statistical filtering on the point cloud to filter out noise, and performing voxel downsampling to reduce the data volume;

[0077] Step S14: performing Euclidean clustering on the merged point cloud and removing small objects;

[0078] Step S15: Remove the ground through normal vector judgment, and then output the remaining point cloud.

[0079] Step S2: Terrain contour calculation.

[0080] Step S21: Create a coordinate point array A with a length of 360, which is used for polar coordinate contour statistics of the point cloud. The point cloud diagram is as follows: Figure 1 As shown, the Z axis is ignored;

[0081] Step S22: traverse the point cloud and convert each point coordinate (x, y, z) into polar coordinate form (l, θ); θ=arctan 2(y,x), the polar coordinate diagram is as follows Figure 2 As shown;

[0082] Step S23: Calculate an index for each point P in the point cloud according to its θ value

[0083] Step S24: Update the contour information of each point P in the point cloud at the index id corresponding to the coordinate point array A:

[0084] If A[id].l>Pl, assign point P to A[id]; the empty coordinate point array A is as follows Figure 3 As shown, after completing step S24, the points screened out are Figure 4 The green points in the middle are assigned to the coordinate point array A;

[0085] Step S25: Connect the points recorded in this array counterclockwise to form a polygon and output it as the contour result, such as Figure 5 As shown;

[0086] Step S3: incremental map update.

[0087] Step S31: In the lanelet2 map, find the mining area or dumping area closest to the current vehicle position. Since the mining area and dumping area are not drivable, their boundaries can be described as a non-drivable polygon, which can be represented by a polygon area with the tag fence in lanelet2. This polygon is denoted as p old ; The contour polygon obtained by the above calculation is recorded as p measure ;

[0088] Step S32: The mining area will become smaller and smaller as the mining progresses; perform geometric Boolean operation p new =p old -p measure ; then use p new Instead of p old To update the map;

[0089] Step S33: The soil discharge area will become larger as the soil discharge progresses. Starting from the current position of the wheel, make a ray from 0°, every 1 degree until 360°. If the ray is aligned with p old and p measure have intersection points, then for p old Do the following: Let the ray and p old The intersection point of the ray is p1, let the ray and p measure The intersection point is p2, execute: p1=p1×0.8+0.2×p2. And replace p with p1 old The nearest point in Figure 6 As shown;

[0090] Working Principle: This embodiment is applied in unmanned mining, specifically for identifying and detecting mining and dumping areas, and automatically updating corresponding high-precision vector maps. This allows maps to reflect the actual terrain in real time, avoiding systemic issues caused by terrain changes caused by mining and dumping operations without corresponding map changes. This method provides a lightweight and efficient method for updating high-precision maps for mining and dumping areas, addressing the low update efficiency of the conventional contour mapping scheme. This embodiment offers simple and efficient computation, allowing a single calculation to update a large map area. The update range is limited to the LiDAR measurement range.

[0091] The rest of this embodiment is the same as that of the above-mentioned embodiment 1, and therefore will not be described in detail.

[0092] Example 3:

[0093] This embodiment, based on any one of the above embodiments 1-2, proposes a vector map update system in an unmanned mine scenario, which is used to execute the above-mentioned vector map update method in an unmanned mine scenario; the system includes a point cloud preprocessing module, a terrain contour calculation module, and a map incremental update module;

[0094] The point cloud preprocessing module is used to preprocess the original point cloud data according to the current posture of the unmanned vehicle;

[0095] The terrain contour calculation module is used to filter dynamic objects and calculate the terrain contour based on the pre-processed original point cloud data;

[0096] The map incremental update module is used to perform Boolean operations on the terrain contour and the high-precision map to obtain an incremental update of the high-precision map.

[0097] This embodiment further proposes an electronic device, including a memory and a processor; the memory stores a computer program; when the computer program is executed on the processor, the above-mentioned vector map update method in the unmanned mine scenario is implemented.

[0098] This embodiment further proposes a computer-readable storage medium having computer instructions stored thereon. When the computer instructions are executed on the above-mentioned electronic device, the above-mentioned vector map updating method in the unmanned mine scenario is implemented.

[0099] The rest of this embodiment is the same as any of the above-mentioned embodiments 1 and 2, and thus will not be described in detail.

[0100] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification or equivalent change made to the above embodiment based on the technical essence of the present invention shall fall within the scope of protection of the present invention.

Claims

1. A vector map updating method in an unmanned mine scenario, characterized in that: First, the raw point cloud data is preprocessed according to the current position of the unmanned vehicle. Then, based on the preprocessed raw point cloud data, dynamic objects are filtered and the terrain contour is calculated. Finally, a Boolean operation is performed on the terrain contour and the high-precision map to obtain an incremental update of the high-precision map.

2. A vector map updating method in an unmanned mine scenario according to claim 1, characterized in that: The vector map updating method in the unmanned mine scenario specifically includes the following steps: Step S1: pre-processing the original point cloud data according to the current position of the unmanned vehicle obtained from the unmanned positioning system; Step S2: Based on the pre-processed original point cloud data, a coordinate point array is established, and the points recorded in the coordinate point array are connected to obtain a contour polygon, which is used as the terrain contour; Step S3: Obtain the area closest to the current vehicle position from the high-precision map, describe the area boundary as an inaccessible polygon, perform Boolean operation on the inaccessible polygon and the outline polygon, and obtain an incremental update of the high-precision map.

3. The vector map updating method in an unmanned mine scenario according to claim 2 is characterized in that: The step S1 specifically includes the following steps: Step S11: obtaining the current position of the unmanned vehicle from the unmanned positioning system; Step S12: transforming the acquired frame of original point cloud data into a world coordinate system according to the current posture of the unmanned vehicle; Step S13: Continuously accumulating multiple frames of point cloud according to a set time period to obtain merged point cloud data; Step S14: performing voxel statistical filtering and voxel downsampling on the merged point cloud data to obtain filtered merged point cloud data; Step S15: performing Euclidean clustering segmentation on the filtered merged point cloud data to obtain merged point cloud data after cluster segmentation; Step S16: removing the ground data from the merged point cloud data after clustering and segmentation according to the normal vector to obtain the pre-processed original point cloud data.

4. The vector map updating method in an unmanned mine scenario according to claim 3 is characterized in that: The step S2 specifically includes the following steps: Step S21: creating a point coordinate array based on the pre-processed original point cloud data; Step S22: traverse the pre-processed original point cloud data and convert the point coordinates into polar coordinates; Step S23: Call the round() function to calculate the index id according to the polar coordinates; Step S24: updating the contour information at the index id corresponding to the coordinate point array; Step S25: Connect the points recorded in the updated coordinate point array in a counterclockwise direction to form a contour polygon, which is used as the terrain contour.

5. The vector map updating method in an unmanned mine scenario according to claim 4 is characterized in that: The step S3 specifically includes the following steps: Step S31: searching the high-precision map for the area closest to the current vehicle location, and describing the area boundary as an inaccessible polygon; the area includes a mining area and a soil dumping area; Step S32: Perform a Boolean operation on the inaccessible polygon and the outline polygon to obtain an incremental update of the high-precision map.

6. The vector map updating method in an unmanned mine scenario according to claim 5 is characterized in that: When the area is a mining area, the specific operation of step S32 is: first, perform geometric Boolean operation on the inaccessible polygon and the outline polygon to obtain P new =P old -P measure Then use P new Instead of P old Update high-precision maps; among them, P measure is the outline polygon, P old A polygon that is not accessible.

7. The vector map updating method in an unmanned mine scenario according to claim 5 is characterized in that: When the area is a soil dumping area, the specific operation of step S32 is: starting from the current vehicle position, starting from 0°, every 1° until 360°, make a ray, if the ray intersects with the inaccessible polygon P old and the contour polygon P measure There are intersection points, then for the inaccessible polygon P old Do the following: Let the ray and P old The intersection point of the ray and the contour polygon P is P1. measure The intersection point is P2, execute: P1=P1×0.8+0.2×P2, and use P1 to replace the inaccessible polygon P old The nearest point.

8. A vector map updating system in an unmanned mine scenario, used to execute the vector map updating method in an unmanned mine scenario according to claim 1; characterized in that: Including point cloud preprocessing module, terrain contour calculation module, map incremental update module; The point cloud preprocessing module is used to preprocess the original point cloud data according to the current posture of the unmanned vehicle; The terrain contour calculation module is used to filter dynamic objects and calculate the terrain contour based on the pre-processed original point cloud data; The map incremental update module is used to perform Boolean operations on the terrain contour and the high-precision map to obtain an incremental update of the high-precision map.

9. An electronic device, characterized in that: It comprises a memory and a processor; a computer program is stored on the memory; when the computer program is executed on the processor, the vector map updating method in the unmanned mine scenario as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions; when the computer instructions are executed on the electronic device according to claim 9, the vector map updating method in the unmanned mine scenario according to any one of claims 1 to 7 is implemented.

Images