A radar calibration method and device based on vehicle trajectory and lane matching

Abstract

The present application provides a radar calibration method and apparatus based on vehicle trajectory and lane matching, relating to the field of radar calibration technology. The method comprises: determining the first pixel position of each pixel point in a target image in a geographic coordinate system based on the lane position; converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system based on the radar position and initial calibration parameters; generating a virtual lane image based on the second vehicle trajectory and gridding parameters; projecting the virtual lane image into the target image based on the second vehicle trajectory and the first pixel position of each pixel point; calculating the deviation between the lane position and the second vehicle trajectory if the size of the overlapping area between the virtual lane image and the target image satisfies a first rule and the gridding parameters satisfy a second rule; and determining the target calibration parameters of the radar to be calibrated based on the virtual lane image when the deviation value is less than a deviation threshold. The present application can achieve efficient and accurate radar calibration.

Description

Technical Field

[0001] The present application relates to the field of radar calibration technology, and in particular to a radar calibration method and device based on vehicle trajectory and lane matching. Background Art

[0002] With the rapid advancement of technology, radar-based road vehicle perception algorithms are increasingly being used in the digital transportation sector. These algorithms can accurately identify and locate vehicles, pedestrians, and other road objects, providing strong support for the construction of precise digital road twin systems. By integrating information such as the location, type, number, and speed of traffic participants, traffic light control systems can achieve more refined control, thereby optimizing intersection efficiency and improving traffic safety. They also provide traffic management departments with real-time, effective traffic monitoring tools, enabling them to quickly respond to traffic accidents and congestion, thereby ensuring road safety.

[0003] LiDAR and millimeter-wave radar, as key sensing devices, can capture real-time information about road targets and their locations. However, this information is presented as distance relative to the radar. To convert radar data into the target's true latitude and longitude, and to achieve target positioning and lane-level matching, the radar equipment must be precisely calibrated to obtain its accurate position and angle. This information can then be calculated based on relative distance to the target's latitude, longitude, and angle. Traditional calibration methods rely on manual labor, requiring not only the extraction of point cloud data in specific scenarios to fit the spatial coordinates of the calibration object, but also the manual matching of targets to lanes, a time-consuming and labor-intensive process.

[0004] In addition, although the existing technology attempts to obtain posture parameters by setting up a marker device with GPS and overlapping it with the point cloud scanned by the lidar, this method also requires on-site operation by personnel and is inefficient. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a radar calibration method and device based on vehicle trajectory and lane matching, so as to solve the problems of high cost and low efficiency of manual calibration in the prior art, and to improve the efficiency and accuracy of radar calibration.

[0006] In a first aspect, the present invention provides a radar calibration method based on vehicle trajectory and lane matching, the method comprising:

[0007] Obtaining initial calibration parameters of the radar to be calibrated, the radar position in the geographic coordinate system, and first vehicle trajectories of each vehicle traveling in the target lane sensed by the radar to be calibrated within a preset time period, as well as a target image including the target lane and the lane position of the target lane in the geographic coordinate system; the first vehicle trajectories are trajectories of the corresponding vehicles in the radar coordinate system;

[0008] Determining a first pixel position of each pixel point in the target image in a geographic coordinate system according to the lane position;

[0009] Converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters;

[0010] mapping the second vehicle trajectory into a pixel grid according to the second vehicle trajectory and configured gridding parameters to generate a virtual lane image;

[0011] projecting the virtual lane image into the target image according to the second vehicle trajectory and the first pixel position of each pixel point;

[0012] If the size of the overlapping area of ​​the virtual lane image and the target image satisfies a configured first rule and the gridding parameters satisfy a configured second rule, calculating a deviation value between the lane position and the second vehicle trajectory;

[0013] When the deviation value is less than a configured deviation threshold, target calibration parameters of the radar to be calibrated are determined based on the virtual lane image.

[0014] In an optional embodiment, the first vehicle trajectory includes the trajectory coordinates of each vehicle at each time point within a preset time period;

[0015] Converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters includes:

[0016] Analyzing the first vehicle trajectory to obtain a vehicle movement direction;

[0017] For any trajectory coordinate, using the initial calibration parameters, converting the radar coordinate of the trajectory coordinate in the radar coordinate system into the Cartesian coordinate of the trajectory coordinate in the geographic coordinate system;

[0018] The radar position and the vehicle movement direction are used to convert the Cartesian coordinates of the trajectory coordinates in the geographic coordinate system into the latitude and longitude coordinates of the trajectory coordinates in the geographic coordinate system.

[0019] In an optional embodiment, the first rule is obtained by matching different configured gridding parameters and different first rules based on the gridding parameters; the gridding parameters include: a grid scale and a spatial boundary; the spatial boundary is determined based on the first pixel position of each pixel point in the target image in the geographic coordinate system;

[0020] Mapping the second vehicle trajectory to a pixel grid according to the second vehicle trajectory and configured gridding parameters to generate a virtual lane image includes:

[0021] For any track point in any second vehicle track, calculating a second pixel position of the track point in the virtual lane image to be generated according to the grid scale and the spatial boundary;

[0022] A virtual lane image is generated based on second pixel positions of different trajectory points in the second vehicle trajectory in the virtual lane image to be generated.

[0023] In an optional embodiment, the size of the overlapping area between the virtual lane image and the target image is determined by calculating the number of target pixels within the overlapping area; the target pixels are pixels corresponding to the target lane and pixels corresponding to the lane line of the target lane;

[0024] The method further comprises:

[0025] If the size of the overlapping area between the virtual lane image and the target image does not satisfy the configured first rule, selecting from the target image an area that does not overlap with the virtual lane image and contains the largest number of target pixels as the target area;

[0026] determining an adjustment parameter based on a relative positional relationship between the target area and the virtual lane image;

[0027] Adjust the initial calibration parameters according to the adjustment parameters to obtain new initial calibration parameters, and return to the execution step of converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters, until the target calibration parameters of the radar to be calibrated are obtained.

[0028] In an optional embodiment, the method further comprises:

[0029] If the gridding parameters do not satisfy the configured second rule, the gridding parameters are adjusted according to the configured adjustment rule to obtain adjusted gridding parameters, the adjusted gridding parameters are used as new gridding parameters, and the process returns to the step of mapping the second vehicle trajectory to a pixel grid according to the second vehicle trajectory and the configured gridding parameters until target calibration parameters for the radar to be calibrated are obtained.

[0030] In an optional embodiment, the method further comprises:

[0031] When the deviation value is not less than the configured deviation threshold, the initial calibration parameters are adjusted according to the difference between the deviation value and the deviation threshold to obtain new initial calibration parameters, and the process returns to the step of converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters, until the target calibration parameters of the radar to be calibrated are obtained.

[0032] In an optional embodiment, after determining the target calibration parameters of the radar to be calibrated, the method further includes:

[0033] When an abnormal signal is received from a radar calibrated using target calibration parameters, the target calibration parameters are used as new initial calibration parameters, and the process returns to the step of converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system based on the radar position and the initial calibration parameters, until new target calibration parameters for the radar are obtained.

[0034] In a second aspect, the present invention provides a radar calibration device based on vehicle trajectory and lane matching, the device comprising:

[0035] an acquisition unit, configured to acquire initial calibration parameters of the radar to be calibrated, the radar position in a geographic coordinate system, and a first vehicle trajectory of each vehicle traveling in a target lane sensed by the radar to be calibrated within a preset time period, as well as a target image including the target lane and the lane position of the target lane in the geographic coordinate system; the first vehicle trajectory being the trajectory of the corresponding vehicle in the radar coordinate system;

[0036] a determining unit, configured to determine a first pixel position of each pixel point in the target image in a geographic coordinate system according to the lane position;

[0037] a conversion unit, configured to convert the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters;

[0038] a generating unit, configured to map the second vehicle trajectory into a pixel grid according to the second vehicle trajectory and configured gridding parameters, to generate a virtual lane image;

[0039] a projection unit, configured to project the virtual lane image into the target image according to the second vehicle trajectory and the first pixel position of each pixel point;

[0040] A calibration unit is configured to calculate a deviation value between the lane position and the second vehicle trajectory if the size of the overlapping area between the virtual lane image and the target image satisfies a configured first rule and the gridding parameters satisfy a configured second rule; and when the deviation value is less than a configured deviation threshold, determine the initial calibration parameters as target calibration parameters for the radar to be calibrated.

[0041] In a third aspect, the present invention provides an electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0042] Memory for storing computer programs;

[0043] The processor is configured to implement any of the methods described in the foregoing embodiments when executing the program stored in the memory.

[0044] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of the aforementioned embodiments is implemented.

[0045] This application uses radar to identify the vehicle trajectories of different vehicles over a period of time, automatically generates virtual lanes similar to map lane lines, and realizes autonomous acquisition of radar posture parameters through matching, fusion analysis, and LCSS deviation analysis with map data. This not only makes up for the shortcomings of existing technologies, but also avoids the problem of recalibration when the equipment shakes or deviates, significantly improving calibration efficiency and accuracy, and injecting new vitality into the development of intelligent transportation systems.

[0046] This application solves the time-consuming and labor-intensive problems of roadside radar pose parameter calibration and radar target mapping to the world coordinate system.

[0047] This application utilizes the device's own sensors and algorithms to automatically complete the calibration process without manual intervention or external equipment assistance. This not only greatly improves the efficiency of calibration, but also reduces errors caused by human factors to a certain extent, thereby improving calibration accuracy.

[0048] This application can reduce dependence on external equipment and professionals, solving the problem that traditional calibration methods often require reliance on expensive measuring equipment and professional technicians. This not only reduces the company's operating costs, but also makes the calibration process more flexible and convenient.

[0049] This application can also reduce dependence on the external environment, such as the impact of environmental factors such as weather, external force, and temperature on the equipment, thereby improving the stability and reliability of the radar perception system; and solve the problem of recalibration due to radar jitter and offset. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 A schematic flow chart of a radar calibration method based on vehicle trajectory and lane matching provided in an embodiment of the present application;

[0052] Figure 2 A schematic diagram of a vehicle trajectory provided in an embodiment of the present application;

[0053] Figure 3 A schematic flow chart of another radar calibration method based on vehicle trajectory and lane matching provided in an embodiment of the present application;

[0054] Figure 4 A schematic diagram of a process for adjusting the overlap area between the second vehicle trajectory and the lane position provided in an embodiment of the present application;

[0055] Figure 5 A schematic flow chart of another radar calibration method based on vehicle trajectory and lane matching provided in an embodiment of the present application;

[0056] Figure 6 A schematic structural diagram of a radar calibration device based on vehicle trajectory and lane matching provided in an embodiment of the present application;

[0057] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0059] The radar calibration method based on vehicle trajectory and lane matching provided in the embodiments of the present application can be applied in a server or a terminal with strong computing capabilities. The server can be a physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. The terminal can be a user equipment (UE) such as a mobile phone, smartphone, laptop, digital broadcast receiver, personal digital assistant (PDA), tablet computer (PAD), handheld device, in-vehicle device, wearable device, computing device, other processing device connected to a wireless modem, mobile station (MS), mobile terminal, etc. The terminal and server can be connected directly or indirectly via wired or wireless communication methods, which is not limited in this application.

[0060] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application can be combined with each other if there is no conflict.

[0061] Figure 1 The flowchart of a radar calibration method based on vehicle trajectory and lane matching provided in the embodiment of the present application is as follows. Figure 1 As shown, the method may include:

[0062] Step S110: Obtain initial calibration parameters of the radar to be calibrated, the radar position in the geographic coordinate system, and the first vehicle trajectory of each vehicle traveling in the target lane sensed by the radar to be calibrated within a preset time period, as well as a target image including the target lane and the lane position of the target lane in the geographic coordinate system.

[0063] Among them, the initial calibration parameters are the initial calibration matrix of the radar; the initial calibration matrix is ​​as follows: , where Tx, Ty, and Tz are the translation amounts in the xyz directions, and are all initially 0. Since the z-axis direction is not adjusted, it can be regarded as ; Radar position is the latitude and longitude coordinates of the installation location of the radar to be calibrated; Lane position is the location of the lane coverage area; The first vehicle trajectory is the trajectory of the corresponding vehicle in the radar coordinate system. The first vehicle trajectory is actually a vehicle trajectory set, {[<x1,t1> ,<x2,t1> .... <x n,t1>][<x1,t2> ,<x2,t2> .... <x n ,t2>]...[ <x1,t m >, <x2,t m >... <x n ,t m >]}, where t m represents the mth time point, x n represents the nth car, [<x1,t1> ,<x2,t1> .... <x n ,t1>] represents the trajectory of each vehicle in the radar coordinate system at time t1; the first vehicle trajectory includes the trajectory coordinates of each vehicle at each time point within a preset time period, and the trajectory coordinates are actually the detection frame coordinates of the detection frame obtained by target detection of each vehicle; the target image can be a vector map, an aerial map or a three-dimensional map; the pixel values ​​corresponding to the pixel points of the target lane in the target image and the pixel values ​​corresponding to the lane line of the target lane are different, so that in the target image, the target lane and the lane line of the target lane are displayed in different colors. For example, in a three-dimensional map, the asphalt road can be set to gray and the lane line can be set to white; on the map after semantic segmentation, the road color can be set to orange, so that the correspondence between the virtual lane and the real map can be more intuitive and accurate; the position of the target lane is determined according to the pixel value of the target lane or the lane line of the target lane, and the lane position of the target lane can be confirmed according to the position of any pixel point in the target image.

[0064] Specifically, due to the possible trajectory changes of a single vehicle target such as lane changing and lane crossing during operation, the real lane cannot be determined based on the vehicle trajectory of a single vehicle; this application increases the richness of trajectory information by obtaining the first vehicle trajectory of each vehicle traveling in the target lane perceived by the radar to be calibrated within a preset time period; obtaining the first vehicle trajectories of multiple vehicles traveling in the target lane, and fully fusing the first vehicle trajectories of multiple vehicles within the preset time period of the same target lane, can accurately reflect the shape of the real lane line.

[0065] Step S120: Determine the first pixel position of each pixel point in the target image in the geographic coordinate system according to the lane position.

[0066] In a specific implementation, based on the lane position and the position of the target lane in the target image, the first pixel position of each pixel point of the target lane in the geographic coordinate system can be determined, and then the first pixel position of each pixel point in the target image in the geographic coordinate system can be determined.

[0067] Step S130: Convert the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters.

[0068] In a specific implementation, the first vehicle trajectory is converted into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters, including:

[0069] The first vehicle trajectory is analyzed to obtain the vehicle movement direction; for any trajectory coordinate, the radar coordinates of the trajectory coordinates in the radar coordinate system are converted into Cartesian coordinates of the trajectory coordinates in the geographic coordinate system using the initial calibration parameters; and the Cartesian coordinates of the trajectory coordinates in the geographic coordinate system are converted into latitude and longitude coordinates of the trajectory coordinates in the geographic coordinate system using the radar position and the vehicle movement direction.

[0070] Specifically, the coordinates of the point cloud data acquired by the radar are relative to the device coordinate system and are distance information from the device. In order for the radar to perceive the vehicle target and ultimately obtain coordinates in the real-world latitude and longitude coordinate system, the coordinates of each target position need to be transformed. Taking the unit vector coordinate conversion of a two-dimensional plane as an example, assuming that the calibration matrix in the two coordinate systems is R, the rotation angle is θ, the coordinates of point p on the road are (x1, y1, z1), and the coordinates in the radar coordinate system are (x2, y2, z2), (x1, y1, z1) can be obtained by the following coordinate system conversion relationship:

[0071] ; Wherein, T represents the deviation value;

[0072] Because the radar is installed horizontally, the z-axis can be ignored. According to the rotation angle θ, x1 and y1 are:

[0073] ;

[0074] We can get:

[0075] so ;

[0076] After the rotation angle, the true latitude and longitude of the target can be calculated based on the radar latitude and longitude coordinates X, Y:

[0077] ;

[0078] From the above, we can see that when the longitude and latitude X, Y and angle θ of the radar position are known, the exact longitude and latitude of the vehicle perceived by the radar can be calculated.

[0079] Step S140: Map the second vehicle trajectory to a pixel grid based on the second vehicle trajectory and the configured gridding parameters to generate a virtual lane image; and project the virtual lane image into the target image based on the second vehicle trajectory and the first pixel position of each pixel point.

[0080] The gridding parameters may include: grid scale and spatial boundary; the spatial boundary is determined based on the first pixel position of each pixel point in the target image in the geographic coordinate system.

[0081] Specifically, mapping the second vehicle trajectory to a pixel grid according to the second vehicle trajectory and the configured gridding parameters to generate a virtual lane image includes:

[0082] For any track point in any second vehicle track, a second pixel position of the track point in the to-be-generated virtual lane image is calculated based on the grid scale and the spatial boundary; and a virtual lane image is generated based on the second pixel positions of different track points in the second vehicle track in the to-be-generated virtual lane image.

[0083] In a specific implementation, the second vehicle trajectory includes the trajectory coordinates of each vehicle at each time point within a preset time period in the geographic coordinate system; however, when generating the virtual lane image, only the trajectory point data in the second vehicle trajectory is used, and the time point data is not required. Therefore, the second pixel positions of different trajectory points in the second vehicle trajectory without time information in the virtual lane image to be generated are fully fused to obtain the virtual lane image, such as Figure 2 As shown in the figure, since the time dimension is removed, the obtained virtual trajectory points will be much denser than the traditional trajectory sequence. The denser the density, the more times the vehicle has traveled, and the closer it is to the real lane trajectory.

[0084] Step S150: Determine target calibration parameters of the radar to be calibrated based on whether the size of the overlapping area between the virtual lane image and the target image meets the configured first rule.

[0085] Among them, the first rule is based on the gridding parameters, and is obtained by matching different configured gridding parameters and different first rules from a comparison table; specifically, the first rule is determined according to the grid scale; when different grid scales are used, the accuracy is different, and the requirements for the size of the overlapping area are different; for example, when the grid scale is larger, the accuracy is lower, and the requirement for the size of the overlapping area is low; when the grid scale is larger, the accuracy is higher, and the requirement for the size of the overlapping area is high; the size of the overlapping area between the virtual lane image and the target image is determined by calculating the number of target pixels in the overlapping area; the target pixel is the pixel corresponding to the target lane and the pixel corresponding to the lane line of the target lane; since the pixel value of the target pixel is different from the pixel value of other pixels, the number of target pixels in the overlapping area can be determined by counting the number of pixel values ​​corresponding to the target pixel on the image.

[0086] In specific implementation, Figure 3 As shown, according to whether the size of the overlapping area of ​​the virtual lane image and the target image meets the configured first rule, the target calibration parameters of the radar to be calibrated are determined, including:

[0087] First, when the size of the overlapping area of ​​the virtual lane image and the target image meets the configured first rule, the target calibration parameters of the radar to be calibrated are determined, including:

[0088] (1) Determine whether the gridding parameters meet the second configuration rule;

[0089] If satisfied, proceed to step (2); otherwise, proceed to step (6);

[0090] (2) Calculating the deviation between the lane position and the second vehicle trajectory using the LCSS algorithm; wherein the deviation may be the deviation between a first virtual centerline of the target lane and a second virtual centerline of the second vehicle trajectory; the first virtual centerline is a virtual line that includes the center point of the target lane and is parallel to the lane line of the target lane; the second virtual centerline is obtained by fitting the second vehicle trajectory of each vehicle; the deviation may also be the deviation between the target lane and the second vehicle trajectory;

[0091] Specifically, the deviation value calculation formula is as follows:

[0092] ;

[0093] Wherein, A represents the second vehicle trajectory; B represents the lane position; in some embodiments of the present application, for convenience of calculation, A may be the second virtual centerline of the second vehicle trajectory; B may be the first virtual centerline of the target lane; ε represents a constant, and in some embodiments of the present application, ε may be 0.5 meters;

[0094] (3) Determine whether the deviation value is less than a configured deviation threshold; when the deviation value between the lane position and the second vehicle trajectory is less than the deviation threshold, it is considered that the lane position and the second vehicle trajectory are consistent, that is, the second vehicle trajectory is accurately mapped, that is, the more similar points the lane position and the second vehicle trajectory have, the more similar the two trajectories are;

[0095] If yes, go to step (4); if no, go to step (5);

[0096] (4) Determine the target calibration parameters of the radar to be calibrated based on the virtual lane image;

[0097] Specifically, the virtual lane image is obtained by mapping the second vehicle trajectory into a pixel grid. Therefore, the position (i.e., longitude and latitude coordinates) of each trajectory point in the virtual lane image can be determined. The longitude and latitude coordinates are determined based on the radar position and the calibration matrix. The radar position remains unchanged. Therefore, the calibration matrix can be determined based on the longitude and latitude coordinates in the virtual lane image.

[0098] (5) According to the difference between the deviation value and the deviation threshold, the initial calibration parameters are adjusted to obtain new initial calibration parameters, and the process returns to step S130;

[0099] (6) According to the configured adjustment rule, the gridding parameter is adjusted to obtain the adjusted gridding parameter, and the adjusted gridding parameter is used as the new gridding parameter, and the process returns to step S130; wherein the adjustment rule is pre-configured by the user, and the adjustment rule is used to characterize the reduction degree of the grid scale; adjusting the gridding parameter means reducing the grid scale according to the configured adjustment rule; in other embodiments of the present application, a gridding parameter set containing multiple gridding parameters can also be pre-configured, and the configured multiple gridding parameters are sorted from large to small according to the grid scale in the gridding parameter set; each time the method of the present application is executed, the gridding parameter ranked first in the gridding parameter set is executed in sequence, that is, the gridding parameter with the largest grid scale is gradually refined to the gridding parameter with the smallest grid scale. By continuously refining and adjusting the gridding parameter, the actual position of the virtual lane can be gradually approached, and the purpose of accurate mapping is finally achieved, which not only improves the accuracy of the mapping, but also makes the entire process more efficient and controllable; when the gridding parameter does not meet the configured second rule, the gridding parameter ranked one place after the current gridding parameter is selected from the gridding parameter set as the adjusted gridding parameter;

[0100] Second, when the size of the overlapping area between the virtual lane image and the target image does not satisfy the configured first rule, determining the target calibration parameters of the radar to be calibrated includes:

[0101] An area of ​​the target image that does not overlap with the virtual lane image and contains the largest number of target pixels is selected as the target area. An adjustment parameter is determined based on the relative positional relationship between the target area and the virtual lane image. The initial calibration parameters are adjusted according to the adjustment parameters to obtain new initial calibration parameters, and the process returns to step S130.

[0102] The adjustment parameters include adjustment methods and adjustment values; the adjustment methods include movement and rotation; and the adjustment value is used to represent the amplitude of movement or rotation.

[0103] Specifically, such as Figure 4 As shown, a target area is selected based on the number of target pixels contained in each area of ​​the target image that does not overlap with the virtual lane image; the virtual lane image is controlled to move toward the target area with the configured adjustment parameters until the virtual lane image covers all target pixels in the target image; and new initial calibration parameters are determined based on the virtual lane image that covers all target pixels in the target image.

[0104] like Figure 5As shown, the radar calibration method based on vehicle trajectory and lane matching provided by the embodiment of the present application is to fuse the vehicle trajectory of each vehicle traveling in the target lane within a preset time period, calculate the deviation from the position of the real lane in the map, and calculate the deviation value (i.e. Figure 4 The size of the difference in the vehicle trajectory is adjusted by continuously subdividing the grid and adjusting the virtual lane to adjust the calibration parameters of the radar used to perceive the vehicle trajectory. This allows the vehicle trajectory to completely match the real lane. The calibration parameters corresponding to the vehicle trajectory when the vehicle trajectory completely matches the real lane are extracted and become the target calibration parameters of the radar.

[0105] In some embodiments, the method further comprises:

[0106] When an abnormal signal is received from a radar calibrated with target calibration parameters, the target calibration parameters are used as new initial calibration parameters, and the execution step is returned to: according to the radar position and the initial calibration parameters, the first vehicle trajectory is converted into a second vehicle trajectory in the geographic coordinate system until the new target calibration parameters of the radar are obtained; wherein, the abnormal signal is a signal output by the radar after jitter or offset due to weather, external force, etc.; when the radar outputs an abnormal signal, its deviation value will change, and the target calibration parameters of the radar obtained before are used as the initial calibration parameters, and the calibration method of the present application is re-executed to achieve automatic adjustment.

[0107] In some embodiments, the radar calibration method based on vehicle trajectory and lane matching further includes:

[0108] Step 1: Data Collection

[0109] Collect all vehicle position data and vehicle perception frame size data perceived by the radar over a period of time.

[0110] Step 2: Obtain the center position information of the lane lines on the map, record it in a sequence, and record the 2D coordinate system corresponding to the map image. Obtain the obvious color characteristics of the lanes based on the image type, for example, the lane colors on a 3D map are gray and white.

[0111] Step 3: Trajectory set formation

[0112] Organize position and size data to form a set of trajectories for different targets at different time points. Remove the time dimension and perform full position fusion to form a virtual lane trajectory. Collect vehicle trajectory data to determine its approximate direction of movement. During the initial run, obtain approximate calibration information (radar latitude and longitude, radar angle) on the map in advance to further speed up the fitting process.

[0113] Step 4: Gridding and Preliminary Mapping

[0114] The virtual lane is divided into grids of different scales, and the largest scale grid is used for the first time to quickly obtain the approximate position parameters.

[0115] Step 5: Color Matching and Adjustment

[0116] The virtual lane trajectory data is projected onto the map image, and the size of the overlapping area between the virtual lane trajectory and the lane position in the target image is determined by calculating the color of the pixel where the trajectory is located;

[0117] Step 6: If the size of the overlapping area does not meet the configured first rule, the virtual lanes are moved and rotated according to the color until all lane colors are covered.

[0118] Step 7: LCSS Deviation Calculation

[0119] The LCSS algorithm is used to calculate the degree of deviation between the virtual lane and the real lane, and the trajectory similarity is evaluated by comparing the distances between trajectory points.

[0120] Step 8: Calibration parameter optimization

[0121] Based on the LCSS score, the grid is continuously subdivided, the virtual lane is adjusted, the LCSS deviation is recalculated, and the calibration parameters are adjusted until the minimum k value is obtained to achieve a complete match between the virtual lane and the real lane.

[0122] Step 9: Monitoring function settings

[0123] Set up monitoring functions to periodically generate and poll virtual lane trajectories and monitor radar device status.

[0124] Step 10: Automatic Repair of Jitter

[0125] When the radar equipment jitters or deflects due to weather, external forces, etc., causing the LCSS value to change beyond the threshold, the mapping deviation automatic adjustment module is restarted, and the previously stored calibration parameter c is used as the initial value for automatic adjustment and repair.

[0126] Corresponding to the above method, the embodiment of the present application also provides a radar calibration device based on vehicle trajectory and lane matching, such as Figure 6 As shown, the radar calibration device based on vehicle trajectory and lane matching includes:

[0127] An acquisition unit 610 is configured to acquire initial calibration parameters of the radar to be calibrated, the radar position in the geographic coordinate system, and first vehicle trajectories of each vehicle traveling in the target lane sensed by the radar to be calibrated within a preset time period, as well as a target image including the target lane and the lane position of the target lane in the geographic coordinate system; the first vehicle trajectory is the trajectory of the corresponding vehicle in the radar coordinate system;

[0128] a determination unit 620, configured to determine a first pixel position of each pixel point in the target image in a geographic coordinate system according to the lane position;

[0129] A conversion unit 630 is configured to convert the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters;

[0130] a generating unit 640 for mapping the second vehicle trajectory into a pixel grid based on the second vehicle trajectory and the configured gridding parameters to generate a virtual lane image;

[0131] a projection unit 650 for projecting the virtual lane image into the target image based on the second vehicle trajectory and the first pixel position of each pixel point;

[0132] A calibration unit 660 is configured to calculate a deviation between the lane position and the second vehicle trajectory if the size of the overlapping area between the virtual lane image and the target image satisfies a configured first rule and the gridding parameters satisfy a configured second rule; and when the deviation is less than a configured deviation threshold, determine the initial calibration parameters as the target calibration parameters for the radar to be calibrated.

[0133] The functions of each functional unit of the radar calibration device based on vehicle trajectory and lane matching provided in the above-mentioned embodiment of the present application can be realized through the above-mentioned method steps. Therefore, the specific working process and beneficial effects of each unit in the radar calibration device based on vehicle trajectory and lane matching provided in the embodiment of the present application will not be repeated here.

[0134] The present application also provides an electronic device, such as Figure 7 As shown, it includes a processor 710 , a communication interface 720 , a memory 730 and a communication bus 740 , wherein the processor 710 , the communication interface 720 , and the memory 730 communicate with each other via the communication bus 740 .

[0135] Memory 730, for storing computer programs;

[0136] The processor 710 is configured to execute the program stored in the memory 730 by performing the following steps:

[0137] Obtaining initial calibration parameters of the radar to be calibrated, the radar position in the geographic coordinate system, and first vehicle trajectories of each vehicle traveling in the target lane sensed by the radar to be calibrated within a preset time period, as well as a target image including the target lane and the lane position of the target lane in the geographic coordinate system; the first vehicle trajectory is the trajectory of the corresponding vehicle in the radar coordinate system;

[0138] According to the lane position, determine the first pixel position of each pixel point in the target image in the geographic coordinate system;

[0139] Converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters;

[0140] mapping the second vehicle trajectory into a pixel grid according to the second vehicle trajectory and the configured gridding parameters to generate a virtual lane image;

[0141] projecting the virtual lane image into the target image according to the second vehicle trajectory and the first pixel position of each pixel point;

[0142] If the size of the overlapping area of ​​the virtual lane image and the target image satisfies the configured first rule and the gridding parameters satisfy the configured second rule, then the deviation value of the lane position and the second vehicle trajectory is calculated;

[0143] When the deviation value is less than the configured deviation threshold, the target calibration parameters of the radar to be calibrated are determined based on the virtual lane image.

[0144] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into address buses, data buses, and control buses. For ease of illustration, the figure uses only one thick line, but this does not mean that there is only one bus or only one type of bus.

[0145] The communication interface is used for communication between the above electronic device and other devices.

[0146] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0147] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0148] The implementation methods and beneficial effects of the various components of the electronic device in the above embodiments to solve the problems can be found in Figure 1 The various steps in the embodiment shown are implemented, therefore, the specific working process and beneficial effects of the electronic device provided by the embodiment of the present application are not repeated here.

[0149] In another embodiment provided in the present application, a computer-readable storage medium is further provided, which stores instructions. When the computer-readable storage medium is executed on a computer, the computer executes the radar calibration method based on vehicle trajectory and lane matching described in any of the above embodiments.

[0150] In another embodiment provided by the present application, a computer program product comprising instructions is also provided. When the computer program product is executed on a computer, the computer executes the radar calibration method based on vehicle trajectory and lane matching described in any one of the above embodiments.

[0151] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the embodiments of the present application may take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware. Furthermore, the embodiments of the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0152] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0153] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0155] Although preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0156] Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. Thus, if these modifications and variations of the embodiments of the present application fall within the scope of the claims and their equivalents, the embodiments of the present application are also intended to include these modifications and variations.

Claims

1. A radar calibration method based on vehicle trajectory and lane matching, characterized in that: The method comprises: Obtaining initial calibration parameters of the radar to be calibrated, the radar position in the geographic coordinate system, and first vehicle trajectories of each vehicle traveling in the target lane sensed by the radar to be calibrated within a preset time period, as well as a target image including the target lane and the lane position of the target lane in the geographic coordinate system; the first vehicle trajectories are trajectories of the corresponding vehicles in the radar coordinate system; Determining a first pixel position of each pixel point in the target image in a geographic coordinate system according to the lane position; Converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters; mapping the second vehicle trajectory into a pixel grid according to the second vehicle trajectory and configured gridding parameters to generate a virtual lane image; projecting the virtual lane image into the target image according to the second vehicle trajectory and the first pixel position of each pixel point; If the size of the overlapping area of ​​the virtual lane image and the target image satisfies a configured first rule and the gridding parameters satisfy a configured second rule, calculating a deviation value between the lane position and the second vehicle trajectory; When the deviation value is less than a configured deviation threshold, target calibration parameters of the radar to be calibrated are determined based on the virtual lane image.

2. The method according to claim 1, wherein The first vehicle trajectory includes the trajectory coordinates of each vehicle at each time point within a preset time period; Converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters includes: Analyzing the first vehicle trajectory to obtain a vehicle movement direction; For any track coordinate, using the initial calibration parameters, converting the radar coordinate of the track coordinate in the radar coordinate system into the Cartesian coordinate of the track coordinate in the geographic coordinate system; The radar position and the vehicle movement direction are used to convert the Cartesian coordinates of the trajectory coordinates in the geographic coordinate system into the latitude and longitude coordinates of the trajectory coordinates in the geographic coordinate system.

3. The method according to claim 1, wherein The first rule is obtained by matching different configured gridding parameters and different first rules based on the gridding parameters; the gridding parameters include: grid scale and spatial boundary; the spatial boundary is determined based on the first pixel position of each pixel point in the target image in the geographic coordinate system; Mapping the second vehicle trajectory to a pixel grid according to the second vehicle trajectory and configured gridding parameters to generate a virtual lane image includes: For any track point in any second vehicle track, calculating a second pixel position of the track point in the virtual lane image to be generated according to the grid scale and the spatial boundary; A virtual lane image is generated based on second pixel positions of different trajectory points in the second vehicle trajectory in the virtual lane image to be generated.

4. The method according to claim 1, wherein The size of the overlapping area between the virtual lane image and the target image is determined by calculating the number of target pixels within the overlapping area; the target pixels are the pixels corresponding to the target lane and the pixels corresponding to the lane line of the target lane; The method further comprises: If the size of the overlapping area between the virtual lane image and the target image does not satisfy the configured first rule, selecting from the target image an area that does not overlap with the virtual lane image and contains the largest number of target pixels as the target area; determining an adjustment parameter based on a relative positional relationship between the target area and the virtual lane image; Adjust the initial calibration parameters according to the adjustment parameters to obtain new initial calibration parameters, and return to the execution step of converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters, until the target calibration parameters of the radar to be calibrated are obtained.

5. The method according to claim 4, wherein The method further comprises: If the gridding parameters do not satisfy the configured second rule, the gridding parameters are adjusted according to the configured adjustment rule to obtain adjusted gridding parameters, the adjusted gridding parameters are used as new gridding parameters, and the process returns to the step of mapping the second vehicle trajectory to a pixel grid according to the second vehicle trajectory and the configured gridding parameters until target calibration parameters for the radar to be calibrated are obtained.

6. The method according to claim 1, wherein The method further comprises: When the deviation value is not less than the configured deviation threshold, the initial calibration parameters are adjusted according to the difference between the deviation value and the deviation threshold to obtain new initial calibration parameters, and the process returns to the step of converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters, until the target calibration parameters of the radar to be calibrated are obtained.

7. The method according to claim 1, wherein After determining the target calibration parameters of the radar to be calibrated, the method further includes: When an abnormal signal is received from a radar calibrated using target calibration parameters, the target calibration parameters are used as new initial calibration parameters, and the process returns to the step of converting the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system based on the radar position and the initial calibration parameters, until new target calibration parameters for the radar are obtained.

8. A radar calibration device based on vehicle trajectory and lane matching, characterized in that: The device comprises: an acquisition unit, configured to acquire initial calibration parameters of the radar to be calibrated, the radar position in a geographic coordinate system, and a first vehicle trajectory of each vehicle traveling in a target lane sensed by the radar to be calibrated within a preset time period, as well as a target image including the target lane and the lane position of the target lane in the geographic coordinate system; the first vehicle trajectory being the trajectory of the corresponding vehicle in the radar coordinate system; a determining unit, configured to determine a first pixel position of each pixel point in the target image in a geographic coordinate system according to the lane position; a conversion unit, configured to convert the first vehicle trajectory into a second vehicle trajectory in a geographic coordinate system according to the radar position and the initial calibration parameters; a generating unit, configured to map the second vehicle trajectory into a pixel grid according to the second vehicle trajectory and configured gridding parameters, to generate a virtual lane image; a projection unit, configured to project the virtual lane image into the target image according to the second vehicle trajectory and the first pixel position of each pixel point; A calibration unit is configured to calculate a deviation value between the lane position and the second vehicle trajectory if the size of the overlapping area between the virtual lane image and the target image satisfies a configured first rule and the gridding parameters satisfy a configured second rule; and when the deviation value is less than a configured deviation threshold, determine the initial calibration parameters as target calibration parameters for the radar to be calibrated.

9. An electronic device, characterized in that: The electronic device includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory for storing computer programs; A processor, configured to implement the method according to any one of claims 1 to 7 when executing a program stored in a memory.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images