An automatic driving control method, system, device and storage medium

Abstract

The embodiment of the application provides a kind of automatic driving control method, system, equipment and storage medium, to solve the technical problems that the existing automatic driving related field engineering research is mainly concentrated in urban road environment, ground automatic driving scheme is basically unusable in mine environment.Method is: receiving the environmental data around the underground vehicle, generates local grid map;The lowest point of the local grid map is shared based on the preset range, to reduce the requirement of detection threshold to echo point number;Wherein, the detection threshold is used to detect remote grid;Introduce intensity threshold to judge the obstacle of remote grid, to build obstacle map;Extract the priori environmental information of the underground vehicle, fuse into the obstacle map, obtain fusion obstacle map;Detect the obstacle existing in the fusion obstacle map, predict the motion state of the obstacle, and generate the driving path of the underground vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, in particular to an automatic driving control method, system, device and storage medium. BACKGROUND

[0002] With the development of artificial intelligence, automatic driving technology has become a hot research topic. Accurate environment perception is the key to automatic driving technology. At present, the construction of ground automatic driving perception system usually adopts one or more of laser radar, millimeter wave radar, ultrasonic radar and camera. Among them, laser radar has always been dominant in the automatic driving perception system due to its high precision, independence on light environment and long detection range. With the development of deep learning, image processing, computer vision and other related fields, cameras are playing an increasingly important role.

[0003] However, for mine environment, underground satellite rejection, variable light, narrow roadway, vehicle explosion-proof and strict safety requirements, the existing ground automatic driving scheme is basically unusable, and the research on automatic driving perception for mine environment is still in the initial stage of theoretical analysis and simulation experiment. First of all, the underground light and color environment is complex, and the computer vision based on camera encounters difficulties. The functions of ultrasonic radar and millimeter wave radar are highly consistent with those of laser radar, and the precision and echo information richness are far inferior to those of laser radar, which makes it difficult to complement laser radar.

[0004] Due to the change of environment, the ground automatic driving perception scheme is not applicable to underground. First of all, the laser radar perception algorithm of traditional ground automatic driving usually uses two types of methods of height difference and scanning line, among which the height difference algorithm has fast calculation speed and low false detection rate, and is more widely used. However, the accurate perception range of the height difference algorithm is limited, and for a 64-line laser radar, the detection range of the oncoming vehicle type obstacle is basically within 30m, so the ground automatic driving often combines with the camera to optimize the perception range, but in the underground environment with poor light and color, such fusion is limited. In addition, accurate obstacle detection, classification and tracking are the key to constructing the automatic driving decision system, but for the automatic driving perception system of single laser radar, the camera lacks the color, contour and other features of the obstacle, and the accuracy of obstacle segmentation, classification and tracking will be greatly reduced. SUMMARY

[0005] The embodiments of the present application provide an automatic driving control method, system, device and storage medium, which solve the technical problem that the existing engineering research in the field of automatic driving is mainly concentrated in urban road environment, and the ground automatic driving scheme cannot be used in mine environment.

[0006] In one aspect, the embodiments of the present application provide an automatic driving control method, which comprises:

[0007] receiving environmental data around a downhole vehicle to generate a local grid map;

[0008] sharing a lowest point of the local grid map based on a preset range to reduce a requirement of a detection threshold on a number of echo points; wherein the detection threshold is used to detect a remote grid;

[0009] introducing an intensity threshold to judge obstacles of the remote grid, thereby constructing an obstacle map;

[0010] extracting prior environmental information of the downhole vehicle and fusing the prior environmental information into the obstacle map to obtain a fused obstacle map;

[0011] detecting obstacles existing in the fused obstacle map, predicting a motion state of the obstacles, and generating a driving path of the downhole vehicle.

[0012] In one implementation of the present application, the process of generating a local grid map specifically comprises:

[0013] receiving echo points of a laser radar to construct an M*N grid map;

[0014] determining a resolution as G and a lower left corner coordinate of the grid as (X_ori, Y_ori), and obtaining a projection of point cloud to grid based on the following transformation,

[0015] row=(x-X ori ) / G

[0016] col=(y-Y ori ) / G

[0017] X ori <x<X ori +M*G

[0018] Y ori <y<Y ori +M*G.

[0019] In one implementation of the present application, the process of extracting prior environmental information of the downhole vehicle and fusing the prior environmental information into the obstacle map to obtain a fused obstacle map specifically comprises:

[0020] generating a prior map based on the prior environmental information;

[0021] obtaining current positioning information of the downhole vehicle, and cutting a part of the obstacle map corresponding to the positioning information;

[0022] screening out the part belonging to the prior map in the cropped map to generate the fusion obstacle map.

[0023] In an implementation manner of the present application, after cropping the part in the obstacle map corresponding to the position of the positioning information, the method further comprises:

[0024] obtaining the angle posture information in the positioning information, rotating the cropped map with the current position as the origin to obtain the coordinate-unified cropped map.

[0025] In an implementation manner of the present application, the detecting the obstacle existing in the fusion obstacle map specifically comprises:

[0026] generating 2D description information of the obstacle points around the underground vehicle;

[0027] performing, on the 2D description information, an abnormal point filtering, a far-end point cloud filtering and a ground segmentation operation in sequence to obtain a 2D grid map;

[0028] performing a point cloud clustering algorithm based on a distance threshold on the obstacle in the 2D grid map to obtain a specific position of the obstacle.

[0029] In an implementation manner of the present application, after obtaining the specific position of the obstacle, the method further comprises:

[0030] determining an occupancy grid of the obstacle in the 2D grid map;

[0031] obtaining a 3D echo point cloud corresponding to the occupancy grid;

[0032] constructing a shape-based classifier based on width and height information of the 3D echo point cloud to classify the obstacle; wherein the types of the obstacle include but are not limited to pedestrians, vehicles and newly accumulated construction materials.

[0033] In an implementation manner of the present application, the predicting the motion state of the obstacle specifically comprises:

[0034] determining the type of the obstacle and historical trajectory information of the obstacle;

[0035] predicting the speed of the obstacle by using Kalman filtering according to the historical trajectory information to obtain a predicted position of the obstacle in a next frame of the fusion obstacle map;

[0036] constructing a ROI region of interest of the obstacle, and performing data association according to the center of the ROI region of interest to realize tracking of the motion state of the obstacle.

[0037] The present application further provides an automatic driving control system, the system comprising:

[0038] an environment data receiving unit configured to receive environment data around a downhole vehicle and generate a local grid map;

[0039] a lowest point sharing unit configured to share a lowest point of the local grid map based on a preset range to reduce a requirement of a detection threshold on a number of echo points; wherein the detection threshold is used to detect a remote grid;

[0040] a strength threshold judging unit configured to introduce a strength threshold to perform obstacle judgment on the remote grid, thereby constructing an obstacle map;

[0041] a fusion unit configured to extract prior environment information of the downhole vehicle and fuse into the obstacle map to obtain a fused obstacle map;

[0042] an obstacle avoidance unit configured to detect an obstacle existing in the fused obstacle map, predict a motion state of the obstacle, and generate a driving path of the downhole vehicle.

[0043] Embodiments of the present application also provide an automatic driving control device, which comprises:

[0044] at least one processor; and

[0045] a memory in communication connection with the at least one processor; wherein

[0046] the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0047] receive environment data around a downhole vehicle and generate a local grid map;

[0048] share a lowest point of the local grid map based on a preset range to reduce a requirement of a detection threshold on a number of echo points; wherein the detection threshold is used to detect a remote grid;

[0049] introduce a strength threshold to perform obstacle judgment on the remote grid, thereby constructing an obstacle map;

[0050] extract prior environment information of the downhole vehicle and fuse into the obstacle map to obtain a fused obstacle map;

[0051] detect an obstacle existing in the fused obstacle map, predict a motion state of the obstacle, and generate a driving path of the downhole vehicle.

[0052] Embodiments of the present application also provide a non-volatile computer storage medium for automatic driving control, which stores computer executable instructions, and the computer executable instructions are configured to:

[0053] receiving environmental data around the underground vehicle to generate a local grid map;

[0054] sharing a lowest point of the local grid map based on a preset range to reduce a requirement of a detection threshold on a number of echo points; wherein the detection threshold is used to detect a remote grid;

[0055] introducing an intensity threshold to judge an obstacle of the remote grid, thereby constructing an obstacle map;

[0056] extracting prior environmental information of the underground vehicle and fusing into the obstacle map to obtain a fused obstacle map;

[0057] detecting an obstacle existing in the fused obstacle map, predicting a motion state of the obstacle, and generating a driving path of the underground vehicle.

[0058] The automatic driving control method, system, device and storage medium provided by the embodiment of the present application improve the ground perception algorithm based on the underground environment characteristics, and adapt to the unique environment and scene in the underground. The traditional height difference algorithm is improved, two characteristics of adjacent grid ground sharing and remote obstacle judgment based on intensity information are added, and the current frame grid map is obtained. Then, the prior map information and the positioning information are fused to construct a complete underground grid map, which provides a perception basis for path planning of the underground autonomous vehicle. Based on the traditional Kalman filtering and ROI (Region of Interest) tracking method, the obstacle type classification is introduced, the ROI interest region size and shape are determined according to the obstacle type, and finally the detection, classification and tracking of the underground pedestrians, vehicles and other obstacles are realized, which provides a perception basis for different driving strategies of the underground autonomous vehicle facing pedestrians and vehicles. The improvement improves the obstacle tracking and prediction accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0059] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0060] Figure 1 The automatic driving control method flowchart provided by the embodiment of the present application;

[0061] Figure 2 The 2D obstacle and grid map provided by the embodiment of the present application;

[0062] Figure 3 The automatic driving control system composition diagram provided by the embodiment of the present application;

[0063] Figure 4A schematic diagram of an automatic driving control device is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0064] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions of the present application will be described below in conjunction with specific embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0065] The embodiments of the present application provide an automatic driving control method, system, device, and storage medium, to solve the technical problem that existing engineering research in the field of automatic driving is mainly concentrated in urban road environment, and a ground automatic driving scheme cannot be used in a mine environment.

[0066] The technical solutions of the embodiments of the present application will be described in detail below with reference to the drawings.

[0067] Figure 1 A flowchart of an automatic driving control method is provided for the embodiments of the present application. As shown in the figure, the method mainly includes the following steps: Figure 1

[0068] Step 101, receiving environmental data around an underground vehicle to generate a local grid map.

[0069] In the embodiments of the present application, first, the echo points of the radar laser are received, and then an M*N grid map is constructed. Second, the resolution of the grid map is determined as G, and the lower left corner coordinates of the grid are (X_ori, Y_ori). Based on the following transformation, the projection of the point cloud to the grid is obtained,

[0070] row = (x - X ori ) / G

[0071] col = (y - Y ori ) / G

[0072] X ori <x < X ori + M*G

[0073] Y ori <y < Y ori + M*G.

[0074] Step 102, sharing the lowest point of the local grid map based on a preset range to reduce the requirement of the detection threshold on the number of echo points; wherein the detection threshold is used to detect the far-end grid.

[0075] ​In the embodiment of the present application, in the traditional height difference algorithm, the farther the laser radar is, the lower the echo point density is, therefore, the detection range based on a single laser radar is limited, and the detection range for the oncoming vehicle is basically within 30 meters. Therefore, in the embodiment of the present application, the lowest points of the local grid map in the preset range are shared, so as to reduce the point requirement of the detection threshold for the echo points, mainly to adapt to the actual situation of low echo point density in the distance, and also to sacrifice the detection accuracy of the height difference algorithm as little as possible. In the embodiment of the present application, only the road points far away are shared with the lowest points of the grid.

[0076] Step 103, introducing an intensity threshold to judge the obstacles of the far-end grid, thereby constructing an obstacle map.

[0077] In the embodiment of the present application, the traditional height difference algorithm does not consider the echo point intensity information, while in the actual mine, the echo point intensity of the laser radar for the far-end vehicle is high, and the echo point intensity of the laser radar for the far-end ground is low, therefore, the intensity threshold is introduced to judge the obstacles of the far-end grid.

[0078] In the embodiment of the present application, after the two optimizations in steps 102 and 103 are realized, the detection range of the 30cm height cone barrel is improved from 15 meters to 30 meters, and the detection range for the oncoming vehicle is improved from 30 meters to 50 meters.

[0079] Step 104, extracting the prior environment information of the underground vehicle, and fusing into the obstacle map to obtain a fused obstacle map.

[0080] In the embodiment of the present application, first, the prior map is generated based on the prior environment information, then the current positioning information of the underground vehicle is obtained, and the part corresponding to the position of the positioning information is cut out in the obstacle map, and the part belonging to the prior map is removed in the cut-out map to generate the obstacle fusion map.

[0081] Specifically, the prior map refers to the map constructed in advance for the automatic driving environment before the automatic driving process is performed. For the underground environment, it basically refers to the roadway walls, construction materials accumulated on both sides of the road and obstacles existing on both sides of the road that affect the driving of other vehicles on the automatic driving route, but the prior map does not contain dynamic pedestrians and vehicles.

[0082] In the embodiment of the present application, after the part corresponding to the positioning information is cut out in the obstacle map, the angle attitude information in the positioning information also needs to be obtained, and then the cut-out map is rotated with the current position as the origin, so as to obtain the coordinate-unified cut-out map.

[0083] Step 105, detecting the obstacles existing in the fusion obstacle map, predicting the motion state of the obstacles, and generating a driving path of the underground vehicle.

[0084] In the embodiments of the present application, first, 2D description information of the obstacle points around the underground vehicle is generated; second, the 2D description information is sequentially subjected to an abnormal point filtering, a far-end point cloud filtering, and a ground segmentation operation to obtain a 2D grid map; and finally, a point cloud clustering algorithm based on a distance threshold is performed on the obstacles in the 2D grid map to obtain specific positions of the obstacles.

[0085] In addition, it is also necessary to determine an occupancy grid of the obstacles in the 2D grid map; then, 3D echo point clouds corresponding to the occupancy grid are acquired; and finally, a shape-based classifier is constructed based on width and height information of the 3D echo point clouds to classify the obstacles; wherein the types of the obstacles include but are not limited to pedestrians, vehicles, and newly accumulated construction materials. As shown in FIG. 2, a local current frame grid map is constructed based on laser radar echo point clouds, and a priori map of the current position is acquired by relying on positioning information; and on the basis of the former, the latter is obtained to acquire a non-priori part of the current frame grid map, which is referred to as a 2D obstacle map. Figure 2

[0086] There are mainly two classification methods for the obstacles, the first one is type classification, which can be divided into pedestrians, vehicles, and others; and the second one is state classification, which can be divided into dynamic and static.

[0087] At present, the first classification mainly relies on the 3D shape of the obstacles. In the detection process, we obtain the occupancy grid corresponding to the 2D obstacles, and according to the grid, we can obtain all the 3D echo point clouds corresponding to the grid. Based on the width and height information of the 3D point clouds, a shape-based classifier can be constructed to classify the vehicle, pedestrian, and other obstacle attributes of the obstacles. Generally, the obstacles with right angles and large ranges are vehicles, the obstacles with a height of 1-2 m and an occupancy grid less than 4 are pedestrians, and the obstacles with a height below 1 m or not meeting other conditions are other obstacles.

[0088] The second classification will be performed after the prediction and tracking of the obstacles. The prediction and tracking will obtain the speed information of the obstacles, and the classifier will classify the dynamic and static obstacles based on the speed threshold. At present, the speed threshold is set to 0.5 m / s.

[0089] ​In the embodiments of the present application, the motion state of the obstacle is predicted, specifically: the type of the obstacle and the historical trajectory information of the obstacle are determined; the speed of the obstacle is predicted by Kalman filtering according to the historical trajectory information, to obtain the predicted position of the obstacle in the next frame in the fused obstacle map; the ROI region of interest of the obstacle is constructed, and data association is performed according to the center of the ROI region of interest, to realize tracking of the motion state of the obstacle.

[0090] Specifically, according to the obstacle classification information and the obstacle position prediction information, the ROI region of interest of the tracked obstacle in the current frame is constructed. The ROI region of interest of the pedestrian is an ellipse constructed with the position in the last frame and the predicted position in the current frame as the foci, and the range of the ellipse is large. The ROI region of interest of the vehicle is a circle constructed with the predicted position in the current frame as the center, and the range of the circle is moderate. The ROI region of interest of other obstacles is classified and discussed according to the historical speed. If the speed is greater than a threshold k, the obstacle is determined as a pedestrian, and if the speed is not greater than the threshold k, the obstacle is determined as a stationary object. A circle is constructed with the position in the last frame as the center, and the range of the circle is small. The currently set threshold k is 0.5 m / s. Data association is performed based on the constructed ROI region of interest and the center (or focus) of the ROI region of interest to realize obstacle tracking.

[0091] The automatic driving control method provided in the embodiments of the present application improves the ground perception algorithm based on the characteristics of the underground environment, and adapts to the unique environment and scene of the underground. The traditional height difference algorithm is improved, two characteristics of adjacent grid ground sharing and remote obstacle judgment based on intensity information are added, and the current frame grid map is obtained; then the prior map information and the positioning information are fused to construct a complete underground grid map, which provides a perception basis for path planning of the underground autonomous vehicle. Based on the traditional Kalman filtering and ROI region of interest tracking method, the obstacle type classification is introduced, the size and shape of the ROI region of interest are determined according to the obstacle type, and finally the detection, classification and tracking of the pedestrians, vehicles and other obstacles in the mine are realized, which provides a perception basis for different driving strategies of the underground autonomous vehicle facing pedestrians and vehicles. The improvement improves the obstacle tracking and prediction accuracy.

[0092] The above is an automatic driving control method provided by the embodiments of the present application, based on the same inventive concept, the embodiments of the present application also provide an automatic driving control system, Figure 3 The automatic driving control system provided in the embodiments of the present application is shown in the composition diagram of the automatic driving control system as shown in Figure 3 The system mainly includes:

[0093] The environment data receiving unit 301 is configured to receive the environment data around the underground vehicle and generate a local grid map.

[0094] The lowest point sharing unit 302 is used to share the lowest point of the local grid map based on a preset range, so as to reduce the requirement of the detection threshold on the number of echo points; wherein, the detection threshold is used to detect distant grids;

[0095] The intensity threshold judgment unit 303 is used to introduce an intensity threshold to judge obstacles in the far-end grid, thereby constructing an obstacle map;

[0096] The fusion unit 304 is used to extract the prior environmental information of the underground vehicle and fuse it into the obstacle map to obtain a fused obstacle map.

[0097] The obstacle avoidance unit 305 is used to detect obstacles in the fused obstacle map, predict the motion state of the obstacles, and generate the driving path of the underground vehicle.

[0098] The above describes an autonomous driving control system provided by an embodiment of this application. Based on the same inventive concept, this application also provides an autonomous driving control device. Figure 4 A schematic diagram of an automatic driving control device provided in this application embodiment is shown below. Figure 4 As shown, the device mainly includes: at least one processor 401; and a memory 402 communicatively connected to the at least one processor. The memory 402 stores instructions executable by the at least one processor 401, which, when executed, enable the at least one processor 401 to: receive environmental data surrounding the underground vehicle and generate a local grid map; share the lowest point of the local grid map based on a preset range to reduce the requirement of the detection threshold on the number of echo points; wherein the detection threshold is used to detect distant grids; introduce an intensity threshold to determine obstacles in the distant grids, thereby constructing an obstacle map; extract the prior environmental information of the underground vehicle and fuse it into the obstacle map to obtain a fused obstacle map; detect obstacles present in the fused obstacle map, predict the movement state of the obstacles, and generate the driving path of the underground vehicle.

[0099] In addition, the embodiment of the present application further provides a nonvolatile computer storage medium for automatic driving control, which stores computer executable instructions, and the computer executable instructions are configured to: receive environmental data around a downhole vehicle, and generate a local grid map; share a lowest point of the local grid map based on a preset range, so as to reduce a requirement of a detection threshold on a number of echo points; wherein the detection threshold is used for detecting a remote grid; introduce an intensity threshold to judge an obstacle of the remote grid, so as to construct an obstacle map; extract prior environmental information of the downhole vehicle, and fuse into the obstacle map to obtain a fused obstacle map; detect an obstacle existing in the fused obstacle map, predict a motion state of the obstacle, and generate a driving path of the downhole vehicle.

[0100] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system) and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The functions specified in one or more flows and / or blocks

[0101] These computer program instructions can also be stored in a computer readable storage medium capable of guiding a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable storage medium produce a product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The functions specified in one or more flows and / or blocks

[0102] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer implemented process, so that the instructions executed on the computer or other programmable device provide a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The functions specified in one or more flows and / or blocks

[0103] In a typical configuration, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memories.

[0104] Each of the embodiments in the present application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each of the embodiments focuses on the difference from other embodiments. In particular, for the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiments.

[0105] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or apparatus that comprises a list of elements does not include only those elements, but can also include other elements not expressly listed or inherent to such process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising the element.

[0106] The above only describes the embodiments of the present application and is not intended to limit the present application. The present application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. An automatic driving control method characterized by comprising: The method comprises: receiving environmental data around the downhole vehicle to generate a local grid map; sharing the lowest point of the local grid map based on a preset range to reduce the requirement of a detection threshold on the number of echo points; wherein the detection threshold is used to detect a remote grid; introducing an intensity threshold to judge obstacles of the remote grid to construct an obstacle map; extracting prior environmental information of the downhole vehicle and fusing it into the obstacle map to obtain a fused obstacle map; specifically, generating a prior map based on the prior environmental information; obtaining current positioning information of the downhole vehicle, and cropping a part of the obstacle map corresponding to the position of the positioning information; screening out the part belonging to the prior map in the cropped map to generate the fused obstacle map; after cropping the part of the obstacle map corresponding to the position of the positioning information, the method further comprises: obtaining angle posture information in the positioning information, rotating the cropped map with the current position as the origin to obtain a coordinate-unified cropped map; detecting obstacles existing in the fused obstacle map, predicting the motion state of the obstacles, and generating a driving path of the downhole vehicle.

2. The automatic driving control method according to claim 1, characterized in that, The process of generating a local grid map specifically comprises: receiving echo points of a laser radar to construct an M*N grid map; determining a resolution G and a lower-left corner coordinate (X_ori, Y_ori) of the grid, and obtaining a projection of point cloud to grid based on the following transformation, row = (x - X ori ) / G col = (y - Y ori ) / G X ori <x<X ori +M*G Y ori <y<Y ori +M*G.

3. The automatic driving control method of claim 1, wherein The detection of obstacles existing in the fused obstacle map specifically comprises: generating 2D description information of obstacle points around the downhole vehicle; performing abnormal point filtering, remote point cloud filtering, and ground segmentation on the 2D description information in sequence to obtain a 2D grid map; performing a distance threshold-based point cloud clustering algorithm on obstacles in the 2D grid map to obtain specific positions of the obstacles.

4. The automatic driving control method according to claim 3, characterized by, After obtaining the specific positions of the obstacles, the method further comprises: determining the occupied grid of the obstacles in the 2D grid map; obtaining 3D echo point cloud corresponding to the occupied grid; constructing a shape-based classifier based on width and height information of the 3D echo point cloud to classify the obstacles; wherein the types of the obstacles include but are not limited to pedestrians, vehicles, and newly accumulated construction materials.

5. The automatic driving control method of claim 1, wherein The prediction of the motion state of the obstacles specifically comprises: determining the type of the obstacles and historical trajectory information of the obstacles; predicting the speed of the obstacles using Kalman filtering based on the historical trajectory information to obtain the predicted position of the obstacles in the next frame of the fused obstacle map; constructing a ROI region of interest of the obstacles, and performing data association based on the center of the ROI region of interest to track the motion state of the obstacles.

6. An automatic driving control system characterized by comprising: The system comprises: an environmental data receiving unit configured to receive environmental data around a downhole vehicle to generate a local grid map; a lowest point sharing unit configured to share the lowest point of the local grid map based on a preset range to reduce the requirement of a detection threshold on the number of echo points; wherein the detection threshold is used to detect a remote grid; The intensity threshold judging unit is configured to introduce an intensity threshold to perform obstacle judgment on the remote grid, so as to construct an obstacle map. The fusion unit is configured to extract prior environmental information of the underground vehicle, and fuse the prior environmental information into the obstacle map to obtain a fused obstacle map. Specifically, the fusion unit is configured to generate a prior map based on the prior environmental information, acquire current positioning information of the underground vehicle, crop a part of the obstacle map corresponding to a position of the positioning information, and filter out the part belonging to the prior map in the cropped map to generate the fused obstacle map. After cropping the part of the obstacle map corresponding to the position of the positioning information, the fusion unit is further configured to acquire angle attitude information in the positioning information, rotate the cropped map with the current position as the origin, and obtain a coordinate-unified cropped map. The obstacle avoidance unit is configured to detect an obstacle existing in the fused obstacle map, predict a motion state of the obstacle, and generate a driving path of the underground vehicle.

7. An automatic driving control device characterized by comprising: The device comprises: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: receive environmental data around an underground vehicle to generate a local grid map; introduce an intensity threshold to perform obstacle judgment on the remote grid, so as to construct an obstacle map; extract prior environmental information of the underground vehicle, and fuse the prior environmental information into the obstacle map to obtain a fused obstacle map; specifically, generate a prior map based on the prior environmental information, acquire current positioning information of the underground vehicle, crop a part of the obstacle map corresponding to a position of the positioning information, and filter out the part belonging to the prior map in the cropped map to generate the fused obstacle map; after cropping the part of the obstacle map corresponding to the position of the positioning information, further include: acquiring angle attitude information in the positioning information, rotating the cropped map with the current position as the origin, and obtaining a coordinate-unified cropped map; detect an obstacle existing in the fused obstacle map, predict a motion state of the obstacle, and generate a driving path of the underground vehicle. The computer executable instructions are configured to: 8.A non-transitory computer storage medium storing computer-executable instructions for automatically driving a vehicle, the computer-executable instructions comprising: receive environmental data around an underground vehicle to generate a local grid map; introduce an intensity threshold to perform obstacle judgment on the remote grid, so as to construct an obstacle map; ​ ​ Extract prior environment information of the underground vehicle, fuse into the obstacle map, and obtain a fused obstacle map; specifically: based on the prior environment information, generate a prior map; obtain current positioning information of the underground vehicle, and crop a part of the obstacle map corresponding to the position of the positioning information; In the cropped map, screen out the part belonging to the prior map to generate the fused obstacle map; after cropping the part of the obstacle map corresponding to the position of the positioning information, further comprising: Obtain angle posture information in the positioning information, rotate the cropped map with the current position as the origin to obtain a coordinate-unified cropped map; Detect the obstacles existing in the fused obstacle map, predict the motion state of the obstacles, and generate a driving path of the underground vehicle.

Images