Target recognition method and system for vehicle, vehicle and device

Abstract

The application discloses a target identification method and system of a vehicle, a vehicle and equipment. The target identification method of the vehicle comprises the following steps: obtaining point cloud data of a radar and perception data of a camera, wherein the radar and the camera are installed on the vehicle; determining an identification area according to the perception data, and filtering the point cloud data according to the identification area to obtain filtered point cloud data; performing target clustering on the filtered point cloud data to obtain clustered targets; obtaining a drivable area according to the perception data, and filtering the clustered targets according to the drivable area and parameters of the clustered targets to obtain a final target identification result. The embodiments of the application can effectively improve the accuracy and reliability of target identification and have high identification efficiency.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, specifically to a vehicle target recognition method, system, vehicle, and device. Background Technology

[0002] Autonomous driving is the trend of automotive development, and its realization involves a number of key technologies, including environmental perception, localization, path planning, and drive-by-wire. Among these, environmental perception is the foundation of the other technologies, and all of these technologies require data obtained from environmental perception.

[0003] Environmental perception technology mainly includes target recognition and state perception. It relies on processing data received by onboard sensors. Considering the complexity of the environment and the limitations of a single sensor, multiple types of sensors are typically used to achieve better perception. Currently, a combination of LiDAR, cameras, and millimeter-wave radar is commonly employed. However, LiDAR is expensive and susceptible to weather conditions, while traditional millimeter-wave radar lacks the ability to measure altitude. 4D millimeter-wave radar, a relatively new sensor, increases the amount of point cloud data and adds altitude information, allowing for more accurate measurement of target state attributes. It is also less expensive and more adaptable. However, regarding target clustering methods using 4D millimeter-wave radar, due to the working principle of millimeter-wave radar, it often detects objects on the road that do not affect vehicle movement, such as manhole covers, leading to false detections and relatively low accuracy and reliability in target recognition. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide a multi-vehicle target recognition method, system, vehicle and equipment that has the advantages of high target recognition accuracy and reliability, and high recognition efficiency.

[0005] In a first aspect, embodiments of this application provide a vehicle target recognition method, including:

[0006] The radar and the camera are mounted on the vehicle.

[0007] The recognition area is determined based on the perception data, and the point cloud data is filtered based on the recognition area to obtain filtered point cloud data.

[0008] The filtered point cloud data is subjected to target clustering to obtain the clustering target;

[0009] The drivable area is obtained based on the perceived data, and the clustered targets are filtered based on the drivable area and the parameters of the clustered targets to obtain the final target recognition result.

[0010] Further, the step of determining the recognition region based on the perceived data and filtering the point cloud data based on the recognition region to obtain filtered point cloud data includes:

[0011] Based on the perceived data, road information on the horizontal plane in the vehicle coordinate system is obtained;

[0012] Based on the road information, the identification area on the horizontal plane is obtained;

[0013] The point cloud data is converted to the vehicle body coordinate system, and the conversion result is mapped onto the horizontal plane.

[0014] The point cloud data corresponding to the mapping results located outside the recognition area on the horizontal plane is removed to obtain the filtered point cloud data.

[0015] Further, obtaining the identification region on the horizontal plane based on the road information includes:

[0016] If a curb is detected, the identification area is obtained based on the curb;

[0017] If no curb is detected but a lane line is detected, the identification area is obtained based on the lane line;

[0018] If no curb or lane line is detected, the recognition area is obtained based on a preset width. Further, after converting the point cloud data to the vehicle coordinate system, the process also includes:

[0019] Point cloud data corresponding to the transformation results located below the horizontal plane are removed.

[0020] Further, the step of performing target clustering on the filtered point cloud data to obtain clustering targets includes:

[0021] The search radius correction coefficient of the point cloud is determined based on the hyperbolic tangent function and the horizontal distance of the point cloud in the vehicle coordinate system in the filtered point cloud data.

[0022] The adaptive search radius of the point cloud is obtained based on the initial search radius and the search radius correction coefficient.

[0023] Obtain the Mahalanobis distance between points in the filtered point cloud data;

[0024] The clustering algorithm based on the adaptive search radius is applied to perform target clustering on the point cloud according to the Mahalanobis distance between the point clouds in the filtered point cloud data, and the clustering target is obtained.

[0025] Further, the step of obtaining the drivable area based on the perceived data, and filtering the clustered targets based on the drivable area and the parameters of the clustered targets to obtain the final target recognition result includes:

[0026] Based on the perceived data, a drivable area is identified, which is an area composed of multiple points in the horizontal plane under the vehicle coordinate system;

[0027] Obtain the height and absolute velocity of the center point of the clustering target;

[0028] For the same clustering objective, the clustering objective is removed when the following conditions are met simultaneously:

[0029] Cluster targets whose projection falls within the drivable area; cluster targets whose center point height is less than a predetermined height; cluster targets whose absolute speed differs from a predetermined speed by less than a preset difference.

[0030] Once the elimination process is complete, the final target identification result is obtained.

[0031] Furthermore, after obtaining the final target recognition result, the process also includes:

[0032] The vehicle is controlled to perform intelligent driving based on the final target recognition result.

[0033] Secondly, embodiments of this application provide a vehicle target recognition system, including:

[0034] An acquisition module is used to acquire point cloud data from a radar and perception data from a camera, wherein the radar and the camera are mounted on a vehicle.

[0035] The first filtering module is used to determine the recognition area based on the perception data, and filter the point cloud data based on the recognition area to obtain filtered point cloud data.

[0036] The clustering module is used to perform target clustering on the filtered point cloud data to obtain clustering targets;

[0037] The second filtering module is used to obtain the drivable area based on the perceived data, and to filter the clustered targets based on the drivable area and the parameters of the clustered targets to obtain the final target recognition result.

[0038] Thirdly, embodiments of this application provide a vehicle, including: a target recognition system for the vehicle according to the second aspect embodiment.

[0039] Fourthly, embodiments of this application provide a computing device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle target recognition method as described in the embodiments of the first aspect of this application.

[0040] The vehicle target recognition method, system, vehicle, and device proposed in this application first filter radar point cloud data based on camera perception data, filtering out some interfering points and point clouds that do not affect vehicle driving, resulting in filtered point cloud data. Then, target clustering is performed on the filtered point cloud data, effectively improving the efficiency and accuracy of clustering. Finally, the clustered targets in the point cloud data are filtered a second time by combining the drivable area detected by the camera. This avoids the radar point cloud data detecting targets on the road that do not affect vehicle driving, such as manhole covers, thus preventing false detections and effectively ensuring the accuracy and reliability of target recognition.

[0041] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0042] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0043] Figure 1 This is a flowchart illustrating a vehicle target recognition method according to an embodiment of this application;

[0044] Figure 2 This is a flowchart illustrating the application of a vehicle target recognition method according to an embodiment of this application.

[0045] Figure 3 This is a structural block diagram of a vehicle target recognition system according to an embodiment of this application;

[0046] Figure 4 A schematic diagram of a computing device suitable for implementing embodiments of this application is shown. Detailed Implementation

[0047] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the relevant application and not intended to limit the application. Furthermore, it should be noted that, for ease of description, only the parts relevant to the application are shown in the accompanying drawings.

[0048] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0049] The following is in conjunction with the appendix Figure 1 This application describes a vehicle target recognition method according to embodiments of the present application, aiming to achieve high target recognition accuracy and efficiency.

[0050] Figure 1 This is a flowchart of a vehicle target recognition method according to an embodiment of this application. Figure 1 As shown, a vehicle target recognition method according to an embodiment of this application includes the following steps:

[0051] S101: Acquire point cloud data from radar and perception data from cameras, wherein the radar and cameras are mounted on the vehicle.

[0052] The radar is a vehicle-mounted radar used to scan point cloud data of the environment in front of and around the vehicle. The camera is a vehicle-mounted camera used to collect image information of the area in front of and around the vehicle; the image information collected by the camera serves as perception data.

[0053] In one embodiment of this application, the radar is, but is not limited to, a 4D millimeter-wave radar. A 4D millimeter-wave radar can obtain more three-dimensional and more abundant point cloud data.

[0054] It should be noted that before obtaining point cloud data through radar and perception data through camera, the 4D millimeter-wave radar and camera should be calibrated and synchronized in time and space. Specifically, the 4D millimeter-wave radar and camera installed on the vehicle should be jointly calibrated to ensure that they are synchronized in time and space.

[0055] After completing the calibration of the 4D millimeter-wave radar and camera, as well as the synchronous processing of time and space, point cloud data is obtained through the 4D millimeter-wave radar, and perception data is obtained through the camera.

[0056] S102: Determine the recognition area based on the perception data, and filter the point cloud data based on the recognition area to obtain filtered point cloud data.

[0057] The identified region can be understood as the region of interest (ROI), that is, the area that needs to be focused on, such as the area that may usually affect vehicle driving.

[0058] In one embodiment of this application, the identification area is determined based on the perception data, and the point cloud data is filtered based on the identification area to obtain filtered point cloud data. Specifically, this may include: obtaining road information on the horizontal plane in the vehicle coordinate system based on the perception data; obtaining the identification area on the horizontal plane based on the road information; converting the point cloud data to the vehicle coordinate system and mapping the conversion result onto the horizontal plane; and removing the point cloud data corresponding to the mapping result located outside the identification area on the horizontal plane to obtain filtered point cloud data.

[0059] In this example, the recognition area on the horizontal plane is obtained based on road information, including: if a roadside is detected, the recognition area is obtained based on the roadside; if no roadside is detected but a lane line is detected, the recognition area is obtained based on the lane line; if neither a roadside nor a lane line is detected, the recognition area is obtained based on a preset width.

[0060] In other words, based on the camera's detection results of the roadside and lane lines, the region of interest (ROI) is determined. Point cloud data falling outside the ROI is removed, thus retaining point cloud data that might affect autonomous driving functions. This reduces the amount of point cloud data, thereby reducing the subsequent computational load related to point cloud data and effectively improving processing efficiency.

[0061] Specifically, the point cloud data from 4D millimeter-wave radar contains some points that are unimportant to the vehicle's current driving state; these are points located outside the Region of Interest (ROI), such as point cloud data at locations far from the sides of the vehicle. To improve the accuracy and efficiency of subsequent point cloud data clustering, the 4D millimeter-wave radar point cloud data undergoes an initial filtering process. After filtering, only point cloud data that may affect the vehicle's driving is retained. The filtering method is as follows:

[0062] The point cloud data from the 4D millimeter-wave radar is transformed into the vehicle coordinate system. Considering the horizontal direction (i.e., within the xOy plane), the results of the roadside and lane lines detected by the camera are obtained. These results represent the curve equations in the horizontal plane (xOy plane) of the vehicle coordinate system, which are then used to determine the ROI (Region of Interest) in the horizontal direction.

[0063] 1. If a roadside edge is detected, then the ROI is the area within the roadside edge;

[0064] 2. If no curb is detected, but two or more lane lines are detected, then the two outermost lane lines are selected, and the ROI is the area within these two lane lines.

[0065] 3. If no curb or sufficient lane markings are detected, the ROI can be determined based on a preset width. This preset width can be determined empirically, for example, set to 6 meters. It should be noted that this 6-meter setting is merely an example. Alternatively, the preset width can be set large enough, for example, greater than the width of the radar's entire horizontal detection area. In this case, the entire horizontal detection area of ​​the radar can be used as the ROI.

[0066] Furthermore, point cloud data often contains interfering points, such as those below the horizontal plane. Therefore, this method, after transforming the point cloud data to the vehicle coordinate system, also includes removing the point cloud data corresponding to the transformed result that is below the horizontal plane. Specifically, the vertical direction (i.e., the z-axis direction) is considered. Taking into account the installation height of the 4D millimeter-wave radar, points below the ground are considered interfering points, and this portion of the radar's point cloud data is removed from the point cloud data. This further improves the accuracy and efficiency of subsequent point cloud data clustering.

[0067] S103: Perform target clustering on the filtered point cloud data to obtain the clustering target.

[0068] By filtering the point cloud data in step S102, some point cloud data that will not affect vehicle driving and some interference points are filtered out, thereby greatly reducing the amount of point cloud data. In this way, when clustering point cloud data using clustering algorithms, the efficiency of clustering can be effectively improved and the amount of computation for clustering can be reduced.

[0069] Among them, some existing clustering algorithms can be used.

[0070] In one embodiment of this application, to improve the clustering effect, the characteristics of 4D millimeter-wave radar point cloud data are fully considered in the clustering of point cloud data. Specifically, the point cloud data acquired by 4D millimeter-wave radar becomes sparser with increasing distance. To achieve good clustering results at both short and long distances, an adaptive search radius is used in this embodiment to achieve clustering. For example, target clustering is performed on filtered point cloud data to obtain clustering targets, including: determining the search radius correction coefficient of the point cloud based on the hyperbolic tangent function (tanh function) and the horizontal distance of the point clouds in the filtered point cloud data in the vehicle coordinate system; obtaining the adaptive search radius of the point cloud based on the initial search radius and the search radius correction coefficient; obtaining the Mahalanobis distance between point clouds in the filtered point cloud data (e.g., calculating the Mahalanobis distance between point clouds based on their coordinates, velocity, etc.); and applying a clustering algorithm based on the adaptive search radius to perform target clustering on the point clouds according to the Mahalanobis distance between the point clouds in the filtered point cloud data to obtain the clustering targets.

[0071] In other words, the Mahalanobis distance between two points is calculated, and the improved DBSCAN clustering algorithm is used to cluster the point cloud data to obtain the clustered targets. Specifically, the point cloud data of the 4D millimeter-wave radar obtained after filtering in step three is clustered. The clustering method can be density clustering based on Mahalanobis distance, such as the improved DBSCAN algorithm.

[0072] For example: using p(x,y,z,v) x ,v y The coordinates of point p are described by x, y, z, and v. x v y The velocity is in the x-axis and y-axis directions.

[0073] Calculate two points p i p j The Mahalanobis distance between them, i.e.:

[0074]

[0075] Since the point cloud data acquired by 4D millimeter-wave radar becomes sparser with increasing distance, an adaptive search radius is adopted to achieve good clustering results at both short and long ranges. Due to the characteristics of the tanh function, it exhibits good matching properties, and its mathematical formula is as follows:

[0076]

[0077] Therefore, the correction coefficient for the search radius is determined based on the tanh function:

[0078]

[0079] in, Let be the horizontal distance from point p. The parameters k, a, and b can be determined through actual testing.

[0080] The initial search radius is set to eps0. The adaptive search radius of each point is calculated as eps = eps0 * f(r). Then, based on the calculated adaptive search radius of each point, the point cloud data of the 4D millimeter-wave radar is clustered to obtain the clustered target result, i.e., the clustered target.

[0081] S104: Obtain the drivable area based on the perception data, and filter the clustered targets based on the parameters of the drivable area and the clustered targets to obtain the final target recognition result.

[0082] Specifically, after clustering is completed, the clustered targets can be filtered again based on the perceived data and the parameters of the clustered targets. The parameters of the clustered targets include, but are not limited to, the height of the center point and the absolute speed of the clustered targets. The final target recognition result can be obtained by: identifying the drivable area based on the perceived data (the drivable area is a region composed of multiple points on the horizontal plane in the vehicle coordinate system); obtaining the height of the center point and the absolute speed of the clustered targets; for the same clustered target, removing the clustered target if the following conditions are met simultaneously: the projection of the clustered target falls within the drivable area; the height of the center point of the clustered target is less than a predetermined height; and the difference between the absolute speed of the clustered target and the predetermined speed is less than a preset difference. After removal, the final target recognition result is obtained.

[0083] Specifically, due to the working principle of millimeter-wave radar, it often detects targets on the road that do not affect vehicle traffic, such as manhole covers, resulting in false detections. Therefore, to further improve the accuracy of target recognition, the clustered targets obtained in step S103 need to be filtered again to achieve more accurate and reliable target detection. For example:

[0084] The system acquires the drivable freespace detected by the camera. This freespace is defined as the area enclosed by several points on the horizontal plane (xOy plane) in the vehicle's coordinate system, denoted as S. For example, the road in front of the vehicle.

[0085] Clustering targets that meet the following filtering conditions are selected: 1. The horizontal projection of the target falls within region S; 2. The height of the target's center point is less than z0; 3. The absolute velocity of the target is less than v0. Here, z0 and v0 are both small thresholds close to 0, which can be determined through actual testing. Clustering targets that simultaneously meet conditions 1, 2, and 3 are removed.

[0086] According to the vehicle target recognition method of this application embodiment, firstly, the radar point cloud data is filtered based on the camera's perception data to remove some interfering points and point clouds that do not affect vehicle driving, resulting in filtered point cloud data. Then, target clustering is performed on the filtered point cloud data, which effectively improves the efficiency and accuracy of clustering. Finally, the clustered targets in the point cloud data are filtered a second time by combining the drivable area detected by the camera. In this way, the radar point cloud data is prevented from detecting targets on the road that do not affect vehicle driving, such as manhole covers, thus avoiding false detection of targets. Therefore, the accuracy and reliability of target recognition are effectively guaranteed.

[0087] like Figure 2 The diagram shown is an application flowchart of the vehicle target recognition method according to an embodiment of this application. Figure 2As shown, the 4D millimeter-wave radar and camera are calibrated. After calibration, point cloud data from the 4D millimeter-wave radar and perception data from the camera are received. Then, the point cloud data is filtered once, and an improved DBSCAN algorithm is used for point cloud clustering. Finally, the clustering results are filtered a second time to obtain and publish the final clustering result, i.e., the target recognition result. This filters the point cloud data from the 4D millimeter-wave radar, retaining data that may affect autonomous driving functions, improving the effectiveness of the clustering results and reducing computational load. Furthermore, compared to existing clustering algorithms, an adaptive search radius is used in the clustering method, improving the accuracy of clustering point clouds at greater distances. Simultaneously, targets that may not affect vehicle movement, such as manhole covers, are further filtered, thereby reducing the false detection rate and improving the accuracy and reliability of target recognition.

[0088] In one embodiment of this application, the vehicle target recognition method, after obtaining the final target recognition result, further includes: controlling the vehicle to perform intelligent driving based on the final target recognition result. That is: the result of step S104 is organized into the final target detection result of the 4D millimeter-wave radar, and then the result can be published to other modules such as data fusion or path planning to support the implementation of related vehicle functions.

[0089] Figure 3 This is a structural block diagram of a vehicle target recognition system according to an embodiment of this application. Figure 3 As shown, the vehicle target recognition system according to an embodiment of this application includes: an acquisition module 310, a first filtering module 320, a clustering module 330, and a second filtering module 340, wherein:

[0090] The acquisition module 310 is used to acquire point cloud data from the radar and perception data from the camera, wherein the radar and the camera are mounted on the vehicle.

[0091] The first filtering module 320 is used to determine the recognition area based on the perception data, and filter the point cloud data based on the recognition area to obtain filtered point cloud data.

[0092] Clustering module 330 is used to perform target clustering on the filtered point cloud data to obtain clustering targets;

[0093] The second filtering module 340 is used to obtain the drivable area based on the perceived data, and filter the clustered targets based on the drivable area and the parameters of the clustered targets to obtain the final target recognition result.

[0094] According to the vehicle target recognition system of this application embodiment, the radar point cloud data is first filtered based on the camera's perception data to remove some interfering points and point clouds that do not affect vehicle driving, resulting in filtered point cloud data. Then, target clustering is performed on the filtered point cloud data, which effectively improves the efficiency and accuracy of clustering. Finally, the clustered targets in the point cloud data are filtered a second time by combining the drivable area detected by the camera. In this way, the radar point cloud data is prevented from detecting targets on the road that do not affect vehicle driving, such as manhole covers, thus avoiding false detection of targets. This effectively ensures the accuracy and reliability of target recognition.

[0095] It should be noted that the specific implementation of the vehicle target recognition system in this application embodiment is similar to the specific implementation of the vehicle target recognition method in this application embodiment. Please refer to the description in the method section for details, which will not be repeated here.

[0096] Furthermore, embodiments of this application provide a vehicle, including a target recognition system for the vehicle according to the above embodiments. The vehicle can filter radar point cloud data based on camera perception data, filtering out some interfering points and point clouds that do not affect vehicle movement, resulting in filtered point cloud data. Then, target clustering is performed on the filtered point cloud data, effectively improving the efficiency and accuracy of clustering. Finally, the clustered targets in the point cloud data are filtered a second time by combining the drivable area detected by the camera. This avoids the radar point cloud data detecting targets on the road that do not affect vehicle movement, such as manhole covers, thus preventing false detections and effectively ensuring the accuracy and reliability of target recognition.

[0097] The following is for reference. Figure 4 , Figure 4 A schematic diagram of a computing device structure suitable for implementing embodiments of this application is shown.

[0098] like Figure 4 As shown, the computer system includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 1002 or programs loaded from storage section 1008 into random access memory (RAM) 1003. RAM 1003 also stores various programs and data required for the system's operating instructions. CPU 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Input / output (I / O) interface 1005 is also connected to bus 1004.

[0099] The following components are connected to I / O interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to I / O interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 1010 as needed so that computer programs read from it can be installed into storage section 1008 as needed.

[0100] Specifically, according to embodiments of this application, the flowchart above refers to... Figure 1 The described process can be implemented as a computer software program. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such an embodiment, the computer program contains program code for performing the methods shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via communication section 1009, and / or installed from removable medium 1011. When the computer program is executed by central processing unit (CPU) 1001, it performs the functions defined in the system of this application.

[0101] It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0102] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operational instructions of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two connected blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified functions or operational instructions, or using a combination of dedicated hardware and computer instructions.

[0103] The units or modules described in the embodiments of this application can be implemented in software or hardware. The described units or modules can also be located in a processor. The names of these units or modules do not, in certain circumstances, constitute a limitation on the unit or module itself.

[0104] In another aspect, this application also provides a computer-readable storage medium, which may be included in the computing device described in the above embodiments, or may exist independently and not assembled into the computing device. The aforementioned computer-readable storage medium stores one or more programs that, when used by one or more processors, execute the vehicle target recognition method described in this application.

[0105] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A vehicle target recognition method, characterized in that, include: The radar and the camera are mounted on the vehicle. The recognition area is determined based on the perception data, and the point cloud data is filtered based on the recognition area to obtain filtered point cloud data. The filtered point cloud data is subjected to target clustering to obtain the clustering target; The drivable area is obtained based on the perceived data, and the clustered targets are filtered based on the drivable area and the parameters of the clustered targets to obtain the final target recognition result. The step of obtaining a drivable area based on the perceived data, and filtering the clustered targets based on the drivable area and the parameters of the clustered targets to obtain the final target recognition result includes: Based on the perceived data, a drivable area is identified, which is an area composed of multiple points in the horizontal plane under the vehicle coordinate system; Obtain the height and absolute velocity of the center point of the clustering target; For the same clustering objective, the clustering objective is removed when the following conditions are met simultaneously: Cluster targets whose projection falls within the drivable area; cluster targets whose center point height is less than a predetermined height; cluster targets whose absolute speed differs from a predetermined speed by less than a preset difference. Once the elimination process is complete, the final target identification result is obtained.

2. The vehicle target recognition method according to claim 1, characterized in that, The step of determining the recognition region based on the perceived data and filtering the point cloud data based on the recognition region to obtain filtered point cloud data includes: Based on the perceived data, road information on the horizontal plane in the vehicle coordinate system is obtained; Based on the road information, the identification area on the horizontal plane is obtained; The point cloud data is converted to the vehicle body coordinate system, and the conversion result is mapped onto the horizontal plane. The point cloud data corresponding to the mapping results located outside the recognition area on the horizontal plane is removed to obtain the filtered point cloud data.

3. The vehicle target recognition method according to claim 2, characterized in that, The step of obtaining the identification area on the horizontal plane based on the road information includes: If a curb is detected, the identification area is obtained based on the curb; If no curb is detected but a lane line is detected, the identification area is obtained based on the lane line; If no curb or lane lines are detected, the recognition area is obtained based on a preset width.

4. The vehicle target recognition method according to claim 2 or 3, characterized in that, After converting the point cloud data to the vehicle coordinate system, the process further includes: Point cloud data corresponding to the transformation results located below the horizontal plane are removed.

5. The vehicle target recognition method according to claim 1, characterized in that, The step of performing target clustering on the filtered point cloud data to obtain clustering targets includes: The search radius correction coefficient of the point cloud is determined based on the hyperbolic tangent function and the horizontal distance of the point cloud in the vehicle coordinate system in the filtered point cloud data. The adaptive search radius of the point cloud is obtained based on the initial search radius and the search radius correction coefficient. Obtain the Mahalanobis distance between points in the filtered point cloud data; The clustering algorithm based on the adaptive search radius is applied to perform target clustering on the point cloud according to the Mahalanobis distance between the point clouds in the filtered point cloud data, and the clustering target is obtained.

6. The vehicle target recognition method according to claim 1, characterized in that, After obtaining the final target recognition result, the following is also included: The vehicle is controlled to perform intelligent driving based on the final target recognition result.

7. A vehicle target recognition system, characterized in that, To implement the vehicle target recognition method according to any one of claims 1-6, comprising: An acquisition module is used to acquire point cloud data from a radar and perception data from a camera, wherein the radar and the camera are mounted on a vehicle. The first filtering module is used to determine the recognition area based on the perception data, and filter the point cloud data based on the recognition area to obtain filtered point cloud data. The clustering module is used to perform target clustering on the filtered point cloud data to obtain clustering targets; The second filtering module is used to obtain the drivable area based on the perceived data, and to filter the clustered targets based on the drivable area and the parameters of the clustered targets to obtain the final target recognition result.

8. A vehicle, characterized in that, include: The vehicle target recognition system according to claim 7.

9. A computing device, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the program, it implements the vehicle target recognition method according to any one of claims 1-6.

Images