Vehicle distance measurement method, device and equipment based on boundary map and storage medium

Abstract

The invention provides a vehicle distance measurement method and device based on a boundary map, equipment and a storage medium, and can be applied to the technical field of autonomous vehicle perception. According to the method, point cloud data are determined through a positioning device and a radar system, algorithm processing is carried out based on the point cloud data to obtain a boundary map, discrete point taking is carried out on the boundary map to obtain a plurality of target interval points, the target interval points are projected to a target detection image detected by a vehicle camera, a plurality of contour projection points are obtained, and a boundary contour is formed. And according to the plurality of contour projection points and the target detection frame position of the to-be-targeted vehicle, distance measurement of the to-be-targeted vehicle can be completed. According to the method, the boundary map information is utilized, so that visual target distance measurement can be accurately and quickly completed.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving vehicle perception technology, and in particular to a vehicle ranging method, device, equipment and storage medium based on a boundary map. Background Art

[0002] Autonomous driving in closed scenarios such as mining areas, ports and industrial parks is one of the important application areas of current industrial intelligence. Autonomous driving technology requires sensors and perception algorithms to achieve environmental perception, and perform path planning and vehicle control based on the perception results to complete the movement of autonomous vehicles.

[0003] Among them, cameras are the most commonly used, lowest-cost and most information-rich sensors, so visual perception is an important component and foundation of autonomous driving perception systems.

[0004] However, existing autonomous driving perception systems are often performed on two-dimensional images, lacking depth information, and after obtaining target detection results, the target needs to be converted into 3D space, which involves the problem of visual ranging.

[0005] Usually, visual ranging is based on the horizon assumption, but this assumption is difficult to meet in some real-world conditions. For example, in the scenario of autonomous driving transportation in mining areas, the vehicle needs to go up and downhill continuously, and the road surface is rugged and the slope changes frequently, which will lead to inaccurate visual ranging. Therefore, a smarter and more accurate vehicle ranging method is needed. Summary of the invention

[0006] The present application provides a vehicle distance measurement method, device, equipment and storage medium based on a boundary map, so as to solve the problem that the existing vehicle distance measurement method is not accurate and fast enough.

[0007] In a first aspect, the present application provides a vehicle ranging method based on a boundary map, the method comprising:

[0008] Determine point cloud data according to the positioning device and the radar system, and perform algorithm processing based on the point cloud data to determine a boundary map;

[0009] Performing discrete point processing on the boundary map to obtain a plurality of target interval points, and performing projection processing on the plurality of target interval points according to a coordinate system conversion relationship between a camera and the radar system to obtain a plurality of contour projection points in a target detection image;

[0010] The distance to the target vehicle is determined according to the multiple contour projection points and the target detection frame position of the target vehicle.

[0011] Optionally, performing algorithm processing based on the point cloud data to determine the boundary contour includes:

[0012] Based on the point cloud data, a boundary map detection is performed to obtain a boundary map detection result, wherein the boundary map detection result is used to indicate first coordinates corresponding to a plurality of boundary points of the boundary map in a radar coordinate system;

[0013] According to the position information of the positioning device, the plurality of first coordinates are converted to obtain the second coordinates of the plurality of boundary points in the world coordinate system, and based on the image detection device, the plurality of second coordinates are converted to obtain the third coordinates of the plurality of boundary points in the image coordinate system;

[0014] The edge map is determined according to the third coordinates of the plurality of boundary points in the image coordinate system.

[0015] Optionally, determining the distance to the target vehicle according to the plurality of contour projection points and the target detection frame position of the target vehicle includes:

[0016] Based on the target detection image, determining a detection frame corresponding to the target vehicle through a target detection algorithm;

[0017] Determining a grounding point corresponding to the target vehicle from the detection frame, the grounding point being used to indicate an estimated position of the target vehicle;

[0018] The distance to the target vehicle is determined according to the multiple contour projection points and the ground contact point.

[0019] Optionally, determining the distance to the target vehicle according to the multiple contour projection points and the grounding point includes:

[0020] Based on the grounding point, determining a target projection point from the multiple contour projection points, the target projection point being a contour projection point that is closest to the grounding point among the multiple contour projection points;

[0021] A target distance between the target projection point and the vehicle is determined, and the target distance is determined as the distance to the target vehicle.

[0022] Optionally, determining the target projection point from the multiple contour projection points based on the ground point includes:

[0023] Determine a grounding line corresponding to the grounding point and a vehicle line corresponding to the vehicle;

[0024] Determine at least one first projection point from the plurality of contour projection points, the first projection point being a projection point between the ground line and the vehicle line;

[0025] For any one of the plurality of first projection points, determining a first distance between the first projection point and the ground point;

[0026] The first projection point corresponding to the minimum first distance is determined as the target projection point.

[0027] Optionally, the step of performing discrete point processing on the boundary contour to obtain a plurality of target interval points includes:

[0028] Determine a plurality of interval discrete parameters according to the vehicle position and a preset interval discrete point picking strategy, wherein the interval discrete parameters are used to indicate corresponding point picking intervals, wherein the preset interval discrete point picking strategy is used to indicate that the point picking intervals between the plurality of interval points on the same side are decreasing;

[0029] According to the plurality of interval discrete parameters, the boundary contour is processed by interval discrete point taking to obtain a plurality of target interval points.

[0030] In a second aspect, the present application provides a vehicle distance measurement device based on a boundary map, the device comprising:

[0031] A determination module, used to determine point cloud data according to the positioning device and the radar system, and perform algorithm processing based on the point cloud data to determine a boundary map;

[0032] A processing module, used for performing interval discrete point processing on the boundary contour to obtain a plurality of target interval points, and performing projection processing on the plurality of target interval points according to a coordinate system conversion relationship between a camera and the radar system to obtain a plurality of contour projection points in the target detection image;

[0033] The determination module is further used to determine the distance to the target vehicle based on the multiple contour projection points and the target detection frame position of the target vehicle.

[0034] Optionally, the processing module is further used to perform boundary map detection based on the point cloud data to obtain a boundary map detection result, wherein the boundary map detection result is used to indicate first coordinates corresponding to multiple boundary points of the boundary map in the radar coordinate system; convert the multiple first coordinates according to the position information of the positioning device to obtain second coordinates of the multiple boundary points in the world coordinate system, and convert the multiple second coordinates based on the image detection device to obtain third coordinates of the multiple boundary points in the image coordinate system;

[0035] The determination module is further used to determine the boundary map according to the third coordinates of the multiple boundary points in the image coordinate system.

[0036] Optionally, the determination module is also used to determine, based on the target detection image and through a target detection algorithm, a detection frame corresponding to the target vehicle; determine a ground point corresponding to the target vehicle from the detection frame, the ground point being used to indicate an estimated position of the target vehicle; and determine the distance to the target vehicle based on the multiple contour projection points and the ground point.

[0037] Optionally, the determination module is also used to determine a target projection point from the multiple contour projection points based on the grounding point, the target projection point being the contour projection point among the multiple contour projection points that is closest to the grounding point; determine a target distance between the target projection point and the vehicle, and determine the target distance as the distance to the target vehicle.

[0038] Optionally, the determination module is also used to determine a grounding line corresponding to the grounding point and a vehicle line corresponding to the vehicle; determine at least one first projection point from the multiple contour projection points, the first projection point being a projection point between the grounding line and the vehicle line; for any one of the multiple first projection points, determine a first distance between the first projection point and the grounding point; and determine the first projection point corresponding to the minimum first distance as the target projection point.

[0039] Optionally, the determination module is further used to determine a plurality of interval discrete parameters according to the vehicle position and a preset interval discrete point picking strategy, wherein the interval discrete parameters are used to indicate corresponding point picking intervals, wherein the preset interval discrete point picking strategy is used to indicate that the point picking intervals between a plurality of interval points on the same side are decreasing;

[0040] The processing module is further used to perform interval discrete point processing on the boundary contour according to the multiple interval discrete parameters to obtain multiple target interval points.

[0041] In a third aspect, the present application provides a vehicle ranging device based on a boundary map, comprising:

[0042] Memory;

[0043] processor;

[0044] Wherein, the memory stores computer-executable instructions;

[0045] The processor executes the computer-executable instructions stored in the memory to implement the vehicle ranging method based on the boundary map as described in the first aspect and various possible implementations of the first aspect.

[0046] In a fourth aspect, the present application provides a computer storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the vehicle ranging method based on a boundary map as described in the first aspect and various possible implementations of the first aspect.

[0047] The present application provides a vehicle ranging method based on a boundary map, which determines point cloud data through a positioning device and a radar system, determines the boundary contour based on the point cloud data, and performs interval discrete point processing on the boundary map to obtain multiple target interval points, and then performs projection processing on the multiple target interval points based on the coordinate system conversion relationship of the camera and radar system to obtain multiple contour projection points in the target detection image, and then based on the grounding point of the target vehicle target detection frame, determines the nearest projection point among the boundary projection points on both sides in front of the grounding point as the position of the target, and compares it with the current position of the vehicle to complete the ranging. This method realizes the ranging of vehicles based on boundary maps, thereby improving the intelligence of autonomous driving technology and effectively preventing accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0049] Figure 1 The process of a vehicle distance measurement method based on a boundary map provided in this application Figure 1 ;

[0050] Figure 2 The process of a vehicle distance measurement method based on a boundary map provided in this application Figure 2 ;

[0051] Figure 3 A schematic diagram of a scene of a vehicle ranging method based on a boundary map provided in this application;

[0052] Figure 4 A schematic diagram of the structure of a vehicle distance measurement device based on a boundary map provided in this application;

[0053] Figure 5 A schematic structural diagram of a vehicle distance measurement device based on a boundary map provided in this application.

[0054] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0056] The terms "first", "second", "third", "fourth", etc. (if any) in the description and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein.

[0057] In the embodiments of the present application, the words "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0058] First, the terms involved in this application are explained:

[0059] Joint calibration: Cameras and LiDAR are important sensors in autonomous driving systems, responsible for collecting images and measuring distances respectively. In order to effectively integrate the data of the two and improve the perception accuracy and stability of the autonomous driving system, they need to be jointly calibrated. The specific steps of this calibration are as follows:

[0060] By moving the calibration plate or vehicle, multiple sets of camera and lidar calibration data are obtained; the images collected by the camera are processed to extract the feature points on the calibration plate; the point cloud data collected by the lidar is processed to extract the point cloud data related to the calibration plate; the spatial mapping relationship between the camera and the lidar is established using the extracted feature points and point cloud data; the extrinsic matrix (including rotation matrix and translation vector) between the camera and the lidar is solved by the optimization algorithm; the point cloud data of the lidar is projected onto the image plane of the camera using the calibrated extrinsic matrix to verify the accuracy of the calibration result; if the projection error is found to be large, it is necessary to recalibrate or adjust the position and angle of the sensor, and apply the calibration result in the autonomous driving system to achieve the fusion of camera and lidar data. Through joint calibration, the spatial alignment and synchronization of the camera and lidar can be ensured, thereby improving the perception accuracy and reliability of the autonomous driving system. This is of great significance for achieving tasks such as autonomous navigation and obstacle avoidance in complex environments.

[0061] Autonomous driving technology: Autonomous driving technology is a type of intelligent connected vehicle technology that uses computer systems to achieve unmanned driving. It relies on artificial intelligence, visual computing, radar, monitoring devices and global positioning systems to work together to enable vehicles to achieve vehicle-road collaboration and automatically and safely operate motor vehicles without human active operation. Autonomous driving vehicles rely on a series of sensors and electronic devices to control the vehicle's driving according to the vehicle's operating status. These sensors include millimeter-wave radar, laser radar and cameras, which work together to accurately identify the vehicle's surroundings, autonomously avoid obstacles in front, and perform automatic steering.

[0062] Autonomous driving in closed scenarios such as mining areas, ports and industrial parks is one of the important application areas of current industrial intelligence. Autonomous driving technology requires sensors and perception algorithms to achieve environmental perception, and perform path planning and vehicle control based on the perception results to complete the movement of autonomous vehicles.

[0063] Among them, cameras are the most commonly used, lowest-cost and most information-rich sensors, so visual perception is an important component and foundation of autonomous driving perception systems.

[0064] However, existing autonomous driving perception systems are often performed on two-dimensional images, lacking depth information, and after obtaining target detection results, the target needs to be converted into 3D space, which involves the problem of visual ranging.

[0065] Usually, visual ranging is based on the horizon assumption, but this assumption is difficult to meet under certain real-world conditions. For example, in autonomous driving transportation scenarios in mining areas, vehicles need to continuously go up and downhill. At the same time, the road surface is rugged and the slope changes frequently, which will lead to inaccurate visual ranging.

[0066] In view of the above problems, this application proposes a vehicle ranging method based on boundary map.

[0067] This method determines point cloud data through positioning equipment and radar system, determines boundary map based on point cloud data, and performs discrete point processing on boundary map to obtain multiple target interval points, and then performs projection processing on multiple target interval points based on the coordinate system conversion relationship of camera and radar system to obtain multiple contour projection points in target detection image, and then determines the nearest projection point among boundary projection points on both sides in front of the grounding point based on the grounding point of target vehicle target detection frame as the position of the target, and compares it with the current position of the vehicle to complete the distance measurement. This method realizes the distance measurement of vehicles based on boundary map, so as to improve the intelligence of autonomous driving technology and effectively prevent accidents.

[0068] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0069] Figure 1 This is a process of a vehicle distance measurement method based on a boundary map provided in an embodiment of the present application. Figure 1 .like Figure 1 As shown, the vehicle distance measurement method based on the boundary map provided in this embodiment includes:

[0070] S101: determining point cloud data according to a positioning device and a radar system, and performing algorithm processing based on the point cloud data to determine a boundary map.

[0071] The positioning device may be, for example, a vehicle positioning device in the prior art, which can detect the position of the vehicle. The radar system may include, for example, a laser radar device or a millimeter wave radar system.

[0072] In one possible approach, map point cloud data can be recorded based on the vehicle positioning device and the radar system, and then the drivable area detection technology can be used to perform boundary map detection on the point cloud data. The boundary map is in the radar coordinate system of the radar system. Combined with the location information of the positioning device, the coordinates corresponding to the boundary map in the boundary map detection result, that is, the radar coordinates, are converted to the world coordinate system, and then the boundary map is determined based on the coordinates in the world coordinate system.

[0073] S102: performing discrete point processing on the boundary map to obtain a plurality of target interval points, and performing projection processing on the plurality of target interval points according to a coordinate system conversion relationship between the camera and the radar system to obtain a plurality of contour projection points in the target detection image.

[0074] Among them, the left and right boundaries in the boundary map under the world coordinate system are processed with discrete points at intervals according to a certain interval, so that multiple target interval points can be obtained.

[0075] For example, when the boundary contour length is 100 meters, discrete points are taken every meter to obtain 101 target interval points. The boundary contour length and the discrete point length are flexibly determined according to actual conditions and are not specifically limited here.

[0076] After obtaining multiple target interval points, since the target interval points are obtained by taking points on the boundary of the boundary map, that is, these multiple target interval points are also in the radar coordinate system, it is necessary to project the multiple target interval points in the radar coordinate system according to the coordinate system conversion relationship between the camera and the radar system, so that these multiple target interval points are projected in the target detection image. At this time, these multiple target interval points become multiple contour projection points, and the multiple contour projection points are connected in series to form the boundary contour in the target detection image.

[0077] It is understandable that this step requires determining the coordinate system of the camera and radar system, and jointly calibrating the camera and lidar to obtain the calibration processing results, and then obtain the posture transformation relationship between the two coordinate systems. The joint calibration process may include, for example: offline joint calibration processing and online joint calibration processing. Offline joint calibration processing refers to recording the data collected during the joint calibration process, which can be subsequently analyzed and processed in a non-real-time environment to complete the initial joint calibration. This method allows in-depth analysis of the calibration data without being restricted by real-time processing, thereby improving the accuracy and reliability of the calibration. The online calibration process uses the offline calibration result as the initial value. The calibration relationship may change due to bumps during the actual operation of the vehicle. Therefore, the online calibration technology can enable it to have real-time correction capabilities to ensure the accuracy of the joint calibration processing results.

[0078] Optionally, the process of joint calibration processing may include, for example: processing the point cloud data collected by the radar system, extracting the point cloud data related to the calibration plate, and using the extracted feature points and point cloud data to establish a spatial mapping relationship between the camera and the lidar; solving the extrinsic parameter matrix (including the rotation matrix and the translation vector) between the camera and the lidar through an optimization algorithm; using the calibrated extrinsic parameter matrix to project the point cloud data of the lidar onto the image plane of the camera, so that the coordinate system transformation relationship between the camera and the radar system can be obtained based on the transformation projection relationship.

[0079] S103: Determine the distance to the target vehicle according to the multiple contour projection points and the target vehicle.

[0080] Among them, the target detection image detected by the camera also includes the target vehicle, and also includes the contour projection points projected by multiple target interval points. Based on the multiple contour projection points and the target vehicle, it is possible to judge and obtain the contour projection point closest to the target vehicle, and use the contour projection point as the position of the target vehicle. At this time, it is only necessary to measure the distance between the contour projection point and the position information obtained by the self-vehicle positioning device to obtain the distance between the self-vehicle and the target vehicle.

[0081] The vehicle ranging method based on the boundary map provided in the embodiment of the present application determines the point cloud data through the positioning device and the radar system, and then performs algorithm processing based on the point cloud data to obtain the boundary map, and then discretely extracts points on the boundary map to obtain multiple target interval points, and projects them into the target detection image detected by the vehicle camera to obtain multiple contour projection points. According to the multiple contour projection points and the target detection frame position of the target vehicle, the distance measurement of the target vehicle can be completed. This method realizes the distance measurement of the vehicle based on the boundary map, so that the visual distance measurement of the target vehicle can be completed accurately and quickly, which has a high improvement effect on the autonomous driving technology.

[0082] Figure 2 The process of the vehicle distance measurement method based on the boundary map provided in the embodiment of the present application Figure 2 This embodiment is in Figure 1 Based on the embodiment, the vehicle distance measurement method based on the boundary map is described in detail. Figure 1 As shown, the vehicle distance measurement method based on the boundary map provided in this embodiment includes:

[0083] S201: Determine point cloud data according to the positioning device and the radar system.

[0084] Among them, step S201 is similar to the above-mentioned step S101, and this application will not repeat them here.

[0085] S202: Perform boundary map detection based on the point cloud data to obtain a boundary map detection result.

[0086] The boundary map detection result is used to indicate first coordinates corresponding to multiple boundary points of the boundary map in the radar coordinate system.

[0087] It can be understood that point cloud data is a series of discrete three-dimensional coordinate points captured by three-dimensional sensors such as lidar, which together depict the three-dimensional structure of the surrounding environment. By processing these point cloud data, accurate detection of object boundaries in target detection images can be achieved. Boundary detection helps the system understand the shape and size of objects, while target vehicle detection provides important information about the location of objects. Combining these two detection results can improve the perception accuracy and safety of the system, enabling the system to better understand and respond to complex traffic scenarios, thereby making more accurate decisions.

[0088] S203: According to the position information of the positioning device, the multiple first coordinates are converted to obtain the second coordinates of the multiple boundary points in the world coordinate system, and based on the image detection device, the multiple second coordinates are converted to obtain the third coordinates of the multiple boundary points in the image coordinate system.

[0089] The boundary point is a radar point of a boundary map detected by a radar. The first coordinate may be, for example, a laser radar coordinate system, the second coordinate may be, for example, a world coordinate system, and the third coordinate may be, for example, an image coordinate system where an image detection device is located.

[0090] In one possible implementation, based on the position information provided by the positioning device, multiple first coordinates in the lidar coordinate system can be first converted into corresponding multiple second coordinates in the world coordinate system, that is, the world coordinates of the boundary points can be obtained; then, using the image detection device, these world coordinates can be further converted into coordinates in the image coordinate system, so as to obtain the exact position of the boundary point in the image.

[0091] S204: Determine the boundary map according to the third coordinates of the plurality of boundary points in the image coordinate system.

[0092] First, the preliminary detection results of the boundary map are obtained through point cloud data analysis, and then the position information provided by the positioning device is used to convert the detected boundary points from the local coordinate system, such as the laser radar coordinate system, to the world coordinate system to obtain the world coordinates of these boundary points. Then, with the help of the image detection device, these world coordinates are further converted to the image coordinate system to obtain the position of the boundary points in the image, that is, the third coordinate. Finally, based on the third coordinates of these boundary points in the image coordinate system, the spatial relationship and connection method between the points are comprehensively considered to construct a complete boundary map.

[0093] It can be understood that through these coordinate points, the location of the boundary map in the image can be more accurately depicted, which is crucial for target recognition, tracking, and subsequent image analysis and processing tasks. This method can improve the accuracy and reliability of target detection and ensure the performance and effect of the image processing system.

[0094] S205: Performing discrete point processing on the boundary map to obtain a plurality of target interval points, and performing projection processing on the plurality of target interval points according to a coordinate system conversion relationship between the camera and the radar system to obtain a plurality of contour projection points in the target detection image.

[0095] Among them, in the process of performing interval discrete point selection on the boundary map, an interval discrete point selection strategy can be preset according to the position of the vehicle. The preset interval discrete point selection strategy can be, for example: the point selection interval between multiple interval points on the same side tends to decrease, that is, when the boundary line of the boundary map closer to the vehicle is used for interval point selection, the larger the point selection interval is, and when the boundary line farther from the vehicle is used for interval point selection, the smaller the point selection interval is. The point selection interval can be represented by an interval discrete parameter, and the two are positively correlated. Therefore, according to multiple interval discrete parameters, the boundary map is processed for interval discrete point selection, and multiple target interval points can be obtained.

[0096] This step can be seen, for example, Figure 3 In the boundary map, the boundary line is processed by discrete points to obtain multiple target interval points. Figure 3 In the figure, when vehicle D is driving, it is necessary to perform visual detection of the road condition ahead. First, the point cloud data of the road condition is determined according to the positioning device and radar system installed in vehicle D, and the boundary map in the road condition is determined according to the point cloud data. Then, the left boundary line in the boundary map is processed by interval discrete point taking to obtain multiple target interval points, and the right boundary line in the boundary map is processed by interval discrete point taking to obtain multiple target interval points.

[0097] It can be understood that after obtaining a plurality of contour projection points, these projection points are connected to form a boundary contour, thereby accurately determining the position and shape of the boundary contour in the image.

[0098] S206: Based on the target detection image, determine a detection frame corresponding to the target vehicle through a target detection algorithm.

[0099] Among them, based on the target detection image, the image can be analyzed using a pre-trained target detection algorithm that can identify specific objects in the image, such as vehicles. Through this process, the algorithm can determine the specific location of the target vehicle in the image and mark it with a detection box for subsequent processing or analysis.

[0100] like Figure 3 As shown in the figure, this is a driving and transportation picture in a closed area scene captured by the camera. The left and right road boundaries are wide near and narrow far away due to the parallel perspective relationship. Between the two road boundaries is the driving road surface, and there are transportation vehicles in the state of meeting and following. In the figure, A is the target detection frame of the meeting vehicle, B and C are the target detection frames of vehicles in the same transportation direction, and a, b, and c are the grounding points of the target detection frames A, B, and C. Projection points 1-25 are projection points projected into the image after discrete points are taken at intervals on the boundary map, of which 1-12 are the left road boundaries and 13-25 are the right road boundaries.

[0101] Optionally, the target vehicle may also be other obstacles, and the obstacles may include, for example, vehicles traveling on the road and other obstacles piled up on the road.

[0102] S207: Determine a grounding point corresponding to the target vehicle from the detection frame.

[0103] The grounding point is used to indicate the estimated position of the target vehicle, and specifically may be the center point position of the lower edge of the target detection frame.

[0104] S208: Determine a grounding line corresponding to the grounding point and a vehicle line corresponding to the vehicle.

[0105] The grounding line may be, for example, a line passing through the grounding point and parallel to the ground, and the vehicle line may be, for example, the center line of the vehicle, which is also parallel to the ground.

[0106] like Figure 3 As shown, taking target vehicle A as an example, a is the grounding point of the target vehicle, the line passing through the grounding point a is the grounding line, and D is the ego vehicle. The target detection image of the camera cannot detect the ego vehicle. Here, an image is specially used for the purpose of expression. The line passing through the origin of the vehicle coordinate system and parallel to the ground is the ego vehicle line.

[0107] S209: Determine at least one first projection point from the multiple contour projection points, where the first projection point is a projection point between the ground line and the vehicle line.

[0108] In order to facilitate detection, when selecting the first projection point, only the projection point between the vehicle line and the ground line can be determined, such as Figure 3 As shown, taking vehicle A as an example, the projection points between the vehicle line and the ground line are 1-6 and 19-25, then the contour projection points of the two parts 1-6 and 19-25 are both the first projection points.

[0109] It is understandable that by determining these first projection points, the results of target detection can be further refined to ensure that the focus is on objects or features that are directly related to vehicle driving safety. For example, in the application of autonomous driving vehicles, these first projection points can help the system identify obstacles on the road ahead, other vehicles or pedestrians and the distance between the vehicle, thereby completing the distance measurement, so that the autonomous driving system can respond in time and take appropriate actions to ensure safe driving.

[0110] S210: For any first projection point among the plurality of first projection points, determine a first distance between the first projection point and the ground point.

[0111] Among them, the Euclidean distance between all the first projection points and the target vehicle is calculated based on the pixel coordinates, that is, the first distance, such as Figure 3As shown, taking vehicle A as an example, the first projection points include contour projection points numbered 1-6 and 19-25. The target vehicle refers to a vehicle that is clearly identified in the image. It is usually a reference vehicle used to calculate the distance to other points. Therefore, it is necessary to calculate the distance between the contour projection points 1-6 and 19-25 and the target vehicle, that is, the distance between the contour projection points 1-6 and 19-25 and the grounding point, and record the distance between each first projection point and the grounding point as the first distance. These first distances will be used for subsequent analysis and comparison to determine which first projection points are closest to the target vehicle.

[0112] It is understandable that the purpose of this step is to evaluate the relative position between each first projection point and the ground point in a quantitative manner so as to further screen out the best target projection point. Such quantitative comparison helps to accurately identify the most suitable one from many candidate points.

[0113] S211: Determine the first projection point corresponding to the minimum first distance as the target projection point.

[0114] In order to determine the specific location of the target projection point, it is first necessary to calculate the first distances between multiple candidate points and the reference point. These candidate points are pre-selected according to a certain rule or algorithm. A first distance value is generated between each pair of candidate points and the reference point. Next, all the calculated first distances are compared. The candidate point with the smallest first distance value will be selected as the final target projection point.

[0115] S212: Determine a target distance between the target projection point and the vehicle, and determine the target distance as the distance to the target vehicle.

[0116] Among them, step S212 is similar to the above-mentioned step S103, and this application will not repeat them here.

[0117] The vehicle ranging method based on the boundary map provided in the embodiment of the present application determines the point cloud data through the positioning device and the radar system, determines the boundary map according to the point cloud data, and performs interval discrete point processing on the boundary map to obtain multiple target interval points, and then performs projection processing on the multiple target interval points according to the coordinate system conversion relationship of the camera and the radar system to obtain multiple contour projection points in the target detection image, and then determines the nearest projection point among the boundary projection points on both sides in front of the grounding point based on the grounding point of the target vehicle target detection frame as the position of the target, and compares it with the current position of the vehicle to complete the ranging. This method realizes the ranging of vehicles based on the boundary map, thereby improving the intelligence of the autonomous driving technology and effectively preventing accidents.

[0118] Figure 4This is a schematic diagram of the structure of the vehicle distance measurement device based on the boundary map provided in this application. Figure 4 As shown, the vehicle distance measurement device 400 based on the boundary map provided in this embodiment includes:

[0119] A determination module 401 is used to determine point cloud data according to the positioning device and the radar system, and perform algorithm processing based on the point cloud data to determine a boundary map;

[0120] The processing module 402 is used to perform interval discrete point processing on the boundary map to obtain a plurality of target interval points, and perform projection processing on the plurality of target interval points according to the coordinate system conversion relationship between the camera and the radar system to obtain a plurality of contour projection points in the target detection image;

[0121] The determination module 401 is further configured to determine the distance to the target vehicle based on the multiple contour projection points and the target detection frame position of the target vehicle.

[0122] Optionally, the processing module 402 is further used to perform boundary map detection based on the point cloud data to obtain a boundary map detection result, where the boundary map detection result is used to indicate first coordinates corresponding to multiple boundary points of the boundary map in the radar coordinate system; convert the multiple first coordinates according to the position information of the positioning device to obtain second coordinates of the multiple boundary points in the world coordinate system, and convert the multiple second coordinates based on the image detection device to obtain third coordinates of the multiple boundary points in the image coordinate system;

[0123] The determination module 401 is further configured to determine the boundary map according to the third coordinates of the plurality of boundary points in the image coordinate system.

[0124] Optionally, the determination module 401 is also used to determine, based on the target detection image and through a target detection algorithm, a detection frame corresponding to the target vehicle; determine a ground point corresponding to the target vehicle from the detection frame, wherein the ground point is used to indicate an estimated position of the target vehicle; and determine a distance to the target vehicle based on the multiple contour projection points and the ground point.

[0125] Optionally, the determination module 401 is also used to determine a target projection point from the multiple contour projection points based on the grounding point, and the target projection point is the contour projection point among the multiple contour projection points that is closest to the grounding point; determine a target distance between the target projection point and the vehicle, and determine the target distance as the distance to the target vehicle.

[0126] Optionally, the determination module 401 is also used to determine a grounding line corresponding to the grounding point and a vehicle line corresponding to the vehicle; determine at least one first projection point from the multiple contour projection points, the first projection point being a projection point between the grounding line and the vehicle line; for any one of the multiple first projection points, determine a first distance between the first projection point and the grounding point; and determine the first projection point corresponding to the minimum first distance as the target projection point.

[0127] Optionally, the determination module 401 is further used to determine a plurality of interval discrete parameters according to the vehicle position and a preset interval discrete point picking strategy, wherein the interval discrete parameters are used to indicate corresponding point picking intervals, wherein the preset interval discrete point picking strategy is used to indicate that the point picking intervals between the plurality of interval points on the same side are decreasing;

[0128] The processing module 402 is further configured to perform interval discrete point processing on the boundary contour according to the multiple interval discrete parameters to obtain multiple target interval points.

[0129] Figure 5 This is a schematic diagram of the structure of the vehicle distance measurement device based on the boundary map provided in this application. Figure 5 As shown, the present application provides a vehicle distance measurement device based on a boundary map. The vehicle distance measurement device 500 based on a boundary map includes: a receiver 501 , a transmitter 502 , a processor 503 and a memory 504 .

[0130] Receiver 501, used for receiving instructions and data;

[0131] A transmitter 502, used for sending instructions and data;

[0132] Memory 504, used to store computer-executable instructions;

[0133] The processor 503 is used to execute the computer-executable instructions stored in the memory 504 to implement the various steps performed by the vehicle distance measurement method based on the boundary map in the above embodiment. For details, please refer to the relevant description in the above embodiment of the vehicle distance measurement method based on the boundary map.

[0134] Optionally, the memory 504 may be independent or integrated with the processor 503 .

[0135] When the memory 504 is independently provided, the electronic device further includes a bus for connecting the memory 504 and the processor 503 .

[0136] The present application also provides a computer storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, a vehicle ranging method based on a boundary map as performed by the vehicle ranging device based on a boundary map is implemented.

[0137] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical nodes; for example, a physical node may have multiple functions, or a function or step may be performed by several physical nodes in cooperation. Some physical nodes or all physical nodes may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or transient medium). As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0138] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0139] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A vehicle distance measurement method based on a boundary map, characterized in that: include: Determine point cloud data according to the positioning device and the radar system, and perform algorithm processing based on the point cloud data to determine a boundary map; Performing discrete point processing on the boundary map to obtain a plurality of target interval points, and performing projection processing on the plurality of target interval points according to a coordinate system conversion relationship between a camera and the radar system to obtain a plurality of contour projection points in a target detection image; The distance to the target vehicle is determined according to the multiple contour projection points and the target detection frame position of the target vehicle.

2. The method according to claim 1, characterized in that The performing algorithm processing based on the point cloud data to determine the boundary map includes: Based on the point cloud data, a boundary map detection is performed to obtain a boundary map detection result, wherein the boundary map detection result is used to indicate first coordinates corresponding to a plurality of boundary points of the boundary map in a radar coordinate system; According to the position information of the positioning device, the plurality of first coordinates are converted to obtain the second coordinates of the plurality of boundary points in the world coordinate system, and based on the image detection device, the plurality of second coordinates are converted to obtain the third coordinates of the plurality of boundary points in the image coordinate system; The boundary map is determined according to the third coordinates of the plurality of boundary points in the image coordinate system.

3. The method according to claim 2, characterized in that The step of determining the distance to the target vehicle according to the plurality of contour projection points and the target detection frame position of the target vehicle comprises: Based on the target detection image, determining a detection frame corresponding to the target vehicle through a target detection algorithm; Determining a grounding point corresponding to the target vehicle from the detection frame, the grounding point being used to indicate an estimated position of the target vehicle; The distance to the target vehicle is determined according to the multiple contour projection points and the ground contact point.

4. The method according to claim 3, characterized in that Determining the distance to the target vehicle according to the plurality of contour projection points and the grounding point comprises: Based on the grounding point, determining a target projection point from the multiple contour projection points, the target projection point being a contour projection point that is closest to the grounding point among the multiple contour projection points; A target distance between the target projection point and the vehicle is determined, and the target distance is determined as the distance to the target vehicle.

5. The method according to claim 4, characterized in that The step of determining a target projection point from the plurality of contour projection points based on the grounding point comprises: Determine a grounding line corresponding to the grounding point and a vehicle line corresponding to the vehicle; Determine at least one first projection point from the plurality of contour projection points, the first projection point being a projection point between the ground line and the vehicle line; For any one of the plurality of first projection points, determining a first distance between the first projection point and the ground point; The first projection point corresponding to the minimum first distance is determined as the target projection point.

6. The method according to claim 3, characterized in that The step of performing discrete point processing on the boundary contour to obtain a plurality of target interval points includes: Determine a plurality of interval discrete parameters according to the vehicle position and a preset interval discrete point picking strategy, wherein the interval discrete parameters are used to indicate corresponding point picking intervals, wherein the preset interval discrete point picking strategy is used to indicate that the point picking intervals between the plurality of interval points on the same side are decreasing; According to the plurality of interval discrete parameters, the boundary contour is processed by interval discrete point taking to obtain a plurality of target interval points.

7. A vehicle distance measuring device based on a boundary map, characterized in that: The device comprises: A determination module, used to determine point cloud data according to the positioning device and the radar system, and perform algorithm processing based on the point cloud data to determine the boundary contour; A processing module, used for performing interval discrete point processing on the boundary contour to obtain a plurality of target interval points, and performing projection processing on the plurality of target interval points according to a coordinate system conversion relationship between a camera and the radar system to obtain a plurality of contour projection points in the target detection image; The determination module is further used to determine the distance to the target vehicle based on the multiple contour projection points and the target detection frame position of the target vehicle.

8. The device according to claim 7, characterized in that The device also includes: The processing module is further used to perform boundary contour detection based on the point cloud data to obtain a boundary contour detection result, wherein the boundary contour detection result is used to indicate first coordinates corresponding to multiple boundary points of the boundary contour in the radar coordinate system; convert the multiple first coordinates according to the position information of the positioning device to obtain second coordinates of the multiple boundary points in the world coordinate system, and convert the multiple second coordinates based on the image detection device to obtain third coordinates of the multiple boundary points in the image coordinate system; The determination module is further configured to determine the boundary contour from the target detection image according to the third coordinates of the multiple boundary points in the image coordinate system.

9. A vehicle distance measuring device based on a boundary map, characterized in that: include: Memory; processor; Wherein, the memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the vehicle distance measurement method based on the boundary map as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores computer execution instructions, which, when executed by a processor, are used to implement the vehicle distance measurement method based on a boundary map as described in any one of claims 1 to 7.

Images