A method, device, equipment and medium for calibrating a radar based on a high-precision map

Abstract

A method for calibrating a radar based on a high-precision map disclosed in this application, the method includes the following steps: obtaining a first historical trajectory of a clustering detection target, forming a topological map according to the first historical trajectory of the clustering detection target, where the clustering detection target refers to a target within the detection range of the radar; comparing the topological map with the actual road in the high-precision map to obtain a first angle between the topological map and the actual road, where the actual road refers to the road corresponding to the first historical trajectory; obtaining a second angle between the actual road and a preset direction; obtaining a third angle between the historical trajectory of the detection target and the preset direction according to the first angle and the second angle; and calibrating the angle of the radar using the third angle.

Description

Technical Field

[0001] This application relates to the field of data processing technologies, and in particular, to a method, apparatus, device, and medium for calibrating a radar based on a high-precision map. Background Art

[0002] To implement a smart road network system, many sensor devices are installed on roadside poles for measuring traffic conditions. Among them, the millimeter-wave radar is one of the most important sensors.

[0003] In the prior art, the detection accuracy of the radar is generally improved by calibration. Traditional calibration methods include: the first is to calibrate the radar with devices such as corner reflectors, and the second is to calibrate the radar by fusing corner reflectors with cameras, lidar, etc. However, since the detection accuracy of the camera deteriorates as the detection distance increases, and the cost of lidar is extremely high, the calibration effect is not ideal enough to fundamentally solve the problem of radar calibration.

[0004] The invention patent with the patent number CN113093128A discloses a radar calibration method. According to the perception result of the millimeter-wave radar on the vehicles driving in the target area, lane topology information is generated in the millimeter-wave radar coordinate system; according to the map data of the target area, lane line topology information is generated in the map coordinate system, and the coordinate system establishment methods of the millimeter-wave radar coordinate system and the map coordinate system are the same; based on a preset deviation amount, the actual parameters for the lane topology information to match the lane line topology information are determined, and the millimeter-wave radar is calibrated using the actual parameters. However, this technical solution defaults that the topological map generated from the radar detection result and the high-precision map Figure 1 are matchable. However, in actual situations, the topological map generated from the radar detection result and the high-precision map are not matchable, and the calibration result is still not accurate enough. Summary of the Invention

[0005] Embodiments of this specification provide a method, apparatus, device, and medium for calibrating a radar based on a high-precision map to solve the problems of inaccurate radar calibration and high cost during calibration.

[0006] Embodiments of this specification adopt the following technical solutions:

[0007] In a first aspect, embodiments of this specification provide a method for calibrating a radar based on a high-precision map, and the method includes the following steps:

[0008] Obtain the first historical trajectory of the clustering detection target, and form a topological map according to the first historical trajectory of the clustering detection target, where the clustering detection target refers to the target within the radar detection range;

[0009] Compare the topological map with the actual road in the high-precision map to obtain a first angle between the topological map and the actual road, where the actual road refers to the road corresponding to the first historical trajectory;

[0010] Obtain a second angle between the actual road and a preset direction;

[0011] Obtain a third angle between the historical trajectory of the detection target and the preset direction according to the first angle and the second angle;

[0012] Use the third angle to calibrate the angle of the radar.

[0013] In a second aspect, an embodiment of this specification also provides a device for calibrating a radar based on a high-precision map, including:

[0014] A trajectory processing module, configured to obtain a first historical trajectory of a clustered detection target and form a topological map according to the first historical trajectory of the clustered detection target, where the clustered detection target refers to a target within the detection range of the radar;

[0015] A comparison module, configured to compare the topological map with the actual road in the high-precision map to obtain a first angle between the topological map and the actual road, where the actual road refers to the road corresponding to the first historical trajectory;

[0016] An extraction module, configured to obtain a second angle between the actual road and a preset direction;

[0017] A calculation module, configured to obtain a third angle between the historical trajectory of the detection target and the preset direction according to the first angle and the second angle;

[0018] A calibration module, configured to use the third angle to calibrate the angle of the radar.

[0019] In a third aspect, an embodiment of this specification also provides an electronic device, including at least one processor and a memory. The memory stores a program and is configured such that at least one processor is used to cause the computer to execute any one of the methods for calibrating a radar based on a high-precision map.

[0020] In a fourth aspect, an embodiment of this specification also provides a computer-readable storage medium, which stores computer instructions for causing the computer to execute any one of the methods for calibrating a radar based on a high-precision map.

[0021] The above at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects: By obtaining the first historical trajectory of the clustering detection target, a topological map is formed according to the first historical trajectory of the clustering detection target; The topological map is compared with the actual road in the high-precision map to obtain the first included angle between the topological map and the actual road; Obtain the second included angle between the actual road and the preset direction; Obtain the third included angle between the historical trajectory of the detection target and the preset direction according to the first included angle and the second included angle; Use the third included angle to calibrate the angle of the radar; Improve the calibration of the detection accuracy of a single radar, realize that on-site personnel do not need to conduct surveying and mapping calibration, and reduce the purpose of calibration costs. Description of the Drawings

[0022] The drawings described herein are used to provide a further understanding of the embodiments of this specification, and constitute a part of the embodiments of this specification. The illustrative embodiments of this specification and their descriptions are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0023] Figure 1 It is a schematic flowchart of a method for calibrating a radar based on a high-precision map provided in Embodiment 1 of this specification;

[0024] Figure 2 It is a schematic diagram of the topological map of the detection target of a method for calibrating a radar based on a high-precision map provided in Embodiment 1 of this specification;

[0025] Figure 3 It is a comparison diagram of the driving trajectory of the detection target and the actual road after calibration of a method for calibrating a radar based on a high-precision map provided in Embodiment 1 of this specification.

[0026] Figure 4 It is a schematic flowchart of a method for calibrating a radar based on a high-precision map provided in Embodiment 2 of this specification;

[0027] Figure 5 It is a schematic diagram of the scene of a method for calibrating a radar based on a high-precision map provided in Embodiment 2 of this specification;

[0028] Figure 6 It is a schematic flowchart of a method for calibrating a radar based on a high-precision map provided in Embodiment 3 of this specification;

[0029] Figure 7 It is a schematic flowchart of a method for calibrating a radar based on a high-precision map provided in Embodiment 4 of this specification;

[0030] Figure 8 It is a two-multiply function projection schematic diagram of a method for calibrating a radar based on a high-precision map provided in Embodiment 4 of this specification;

[0031] Figure 9Schematic diagram of a device structure for calibrating a roadside millimeter-wave radar based on a high-precision map provided for Embodiment 5. Detailed implementation manners

[0032] In the prior art, in order to implement an intelligent road network system, many sensor devices are installed on roadside poles for measuring traffic conditions. Among them, the millimeter-wave radar is one of the most important sensors.

[0033] To avoid interference with the 77G frequency band of in-vehicle millimeter-wave radars, most of the installation standards for roadside devices are millimeter-wave radars with a 24G frequency band. However, the 24G millimeter-wave radar has a relatively large lateral detection accuracy error. It is not easy to jointly calibrate the radar detection results at multiple point positions on the poles, and manual adjustment is required, which consumes costs and has a low calibration accuracy.

[0034] In the prior art, calibration methods are generally used to improve the detection accuracy of radars. Traditional calibration methods include: First, using devices such as corner reflectors to calibrate the radar, and second, calibrating the radar by fusing corner reflectors with cameras, lidar, etc. However, since the detection accuracy of cameras becomes worse as the detection distance increases, and the cost of lidar is extremely high, the calibration effect is not ideal enough and cannot fundamentally solve the problem of radar calibration.

[0035] Therefore, the embodiments of this specification provide a method for calibrating a radar based on a high-precision map. By obtaining the first historical trajectory of a clustering detection target, a topological map is formed according to the first historical trajectory of the clustering detection target; the topological map is compared with the actual road in the high-precision map to obtain the first included angle between the topological map and the actual road; the second included angle between the actual road and a preset direction is obtained; the third included angle between the historical trajectory of the detection target and the preset direction is obtained according to the first included angle and the second included angle; the radar is calibrated using the third included angle; the calibration of the detection accuracy of a single radar is improved, and the purpose of realizing the calibration without on-site personnel surveying and mapping and reducing the calibration cost is achieved.

[0036] To make the purpose, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0037] The following will detail the technical solutions provided by each embodiment of this specification in conjunction with the drawings.

[0038] Embodiment 1

[0039] For a roadside radar, the most difficult thing to calculate is the orientation of the radar. During the installation and construction of the radar, the workers install the radar randomly facing the road surface. Therefore, the calibration personnel do not know the angle between the installation angle of the radar and the due north direction. The radar detection target position information only has a relative position relationship with the radar, and the GPS value of the target absolute position information cannot be calculated. Only by correctly obtaining the angle value between the radar detection wave emission direction and the due north direction can the GPS value of the detected object be correctly calculated. Since the number of roadside radars is huge, if workers are required to survey and map while installing on-site, it will cost a huge amount. Embodiment 1 provides a method for calibrating a radar based on a high-precision map, which can solve the problem that calibration personnel can accurately obtain the radar orientation information and calibrate the radar without going to the site for surveying and mapping. Please refer to Figure 1 as shown in Figure 1 FIG. 4 is a schematic flowchart of a method for calibrating a radar based on a high-precision map provided in Embodiment 1 of this specification.

[0040] Specifically, the method includes the following steps:

[0041] S101. Obtain the first historical trajectory of the clustering detection target, and form a topological map according to the first historical trajectory of the clustering detection target, where the clustering detection target refers to the target within the detection range of the radar.

[0042] Specifically, the technical solution listed in this embodiment preferentially calibrates the radar installed on the roadside, but in specific applications, it is not limited where the radar is installed. In this embodiment, the clustering detection target is exemplified by a normal sedan. After the roadside radar detects the clustering detection target, it uploads the historical trajectory of the clustering detection target's travel to the server. The server forms a topological map according to the historical trajectory of the clustering detection target. The method of forming a topological map from the historical trajectory of the clustering detection target can adopt ordinary technical means. Please refer to Figure 2 as shown in Figure 2 FIG. 5 is a schematic diagram of the topological map. The data transmission method includes but is not limited to wired, wireless and other methods, and the transmission method is selected according to needs and is not limited here.

[0043] It should be noted that before obtaining the first historical trajectory of the radar's detection of clustered detection targets, since the environment of the roadside device is in a fixed state, a large amount of fixed environmental information can be used to guide the detection results, eliminate unnecessary information, and ensure the complete accuracy of the detection. Specifically, by setting the positions of the sidewalk and the stop line, the radar is prevented from filtering out stationary objects at these locations. Since the radar can only detect moving objects, stationary objects will be directly filtered out by the radar as noise interference signals. In this way, vehicles stopped at intersections will be directly blocked. Therefore, by presetting the position of the sidewalk stop line in the radar, detection targets that change from moving to stationary at this location can be retained, enabling the radar to have an estimate of the detection targets that change from moving to stationary. After the detection target moves again and is detected by the radar, it will not be misidentified as another object, thus achieving the purpose of maintaining a constant ID for the radar's tracking of the detection target and reducing interference.

[0044] Preferably, due to the interference of uncertain factors such as the movement of trees caused by wind and flying birds, on straight roads, detection targets with significantly different lane line directions are blocked. Exceptionally, at intersections, since the movement directions of pedestrians are irregular, small targets moving in unconventional directions at these locations are not filtered out.

[0045] More preferably, by setting up an electronic fence, special processing is performed on targets within different electronic fences, and objects at positions of no interest are blocked.

[0046] Furthermore, it should be noted that although the radar has high detection accuracy at close range, its detection resolution decreases as the distance increases. Therefore, the radar may misjudge some large targets. For example, for a truck, the radar may misjudge it as multiple small cars and generate multiple detection points. In this embodiment, according to the width of the lane line and lane information, clustering is performed on the reflected targets. Specifically, the distance between two adjacent distributed detection targets is obtained, and it is determined whether the distance exceeds a preset length. If it does not exceed, the two adjacent distributed detection targets are clustered into the same type of target. The preset length can be the width of the lane line. When the distance between two adjacent detection points does not exceed the width of the lane line of the lane where the distributed detection target is located, the two detection points are regarded as the same detection target, forming a clustered detection target.

[0047] For example, the target of misidentifying a large truck as multiple small cars is reclustered into a large truck to improve the accuracy of the data.

[0048] It should be understood that the above-listed relevant specific contents are only for illustrative purposes and should not impose any limitations on the present invention.

[0049] S103. Compare the topological map with the actual road in the high-precision map to obtain the first included angle between the topological map and the actual road, where the actual road refers to the road corresponding to the first historical trajectory.

[0050] Specifically, please continue to refer to Figure 2 As shown, the topological map generated from the driving trajectory of the clustering detection target (hereinafter referred to as the detection target) is in the shape of a linear structure composed of several points, such as Figure 2 the topological map of the detection target composed of points in Figure 2 The thin solid line in is the actual road in the high-precision map. By comparing the topological map with the actual road in the high-precision map, the first included angle between the topological map and the actual road can be obtained.

[0051] S105. Obtain the second included angle between the actual road and the preset direction.

[0052] Specifically, the preset direction includes but is not limited to the due north direction, due south direction, due east direction, and due west direction. In this embodiment, the due north direction is adopted and can be selected according to needs in specific applications, which is not limited here. The included angle between the actual road and the due north direction in the high-precision map is stored in the data memory library, so it can be directly obtained in the server.

[0053] S107. Obtain the third included angle between the historical trajectory of the detection target and the preset direction according to the first included angle and the second included angle.

[0054] Specifically, if both the first included angle and the second included angle contain positive and negative signs in front, then add the first included angle and the second included angle to obtain the included angle between the historical trajectory of the detection target and the due north direction. If neither the first included angle nor the second included angle contains positive and negative signs in front, then add or subtract the first included angle and the second included angle to obtain the included angle between the historical trajectory of the detection target and the due north direction, depending on the specific situation.

[0055] S109. Calibrate the angle of the radar using the third included angle.

[0056] Specifically, calibrating the angle of the radar using the third included angle can be understood as: inputting the third included angle between the historical trajectory of the detection target and the due north direction into the radar driver program to calibrate the radar, realizing rapid deployment and calibration, and the operation method of inputting into the radar driver program to calibrate the radar can adopt the ordinary method. Please refer to Figure 3 as shown Figure 3 is a comparison diagram of the driving trajectory of the detection target and the actual road after calibration. After calibration, the radar detection target can be converted into absolute GPS coordinate values.

[0057] This embodiment can form a topological map based on the historical trajectory of the detection target by obtaining the historical trajectory of the detection target, where the detection target refers to the target within the radar detection range; compare the topological map with the actual road in the high-precision map to obtain the first angle between the topological map and the actual road, where the actual road refers to the road on the high-precision map, and the road is the road on which the detected target generates the historical trajectory after driving; obtain the second angle between the actual road and the due north direction; add the first angle and the second angle to obtain the angle between the historical trajectory of the detection target and the due north direction; input the angle between the historical trajectory of the detection target and the due north direction into the radar for calibration; improve the calibration of the detection accuracy of a single radar, achieve the purpose of not requiring on-site personnel for surveying and calibration, and reduce the calibration cost; at the same time, cluster radar targets according to the lane line information of the high-precision map, reduce interference, improve the accuracy of calibration, and guide the detection results through a large amount of fixed environmental information, eliminate unnecessary information, and ensure the complete accuracy of detection.

[0058] It should be understood that the above-listed relevant specific contents are only for illustrative purposes and should not impose any limitations on the present invention.

[0059] Embodiment 2

[0060] After the radar is calibrated, the radar detection target can be converted into an absolute GPS coordinate value. However, on roads with a large bend radius, it is not possible to directly overlap the topological map drawn by the radar detection object and the high-precision map. And for 24G single-radar detection, the angular resolution is only 6 degrees, so there will be a situation where the object shakes horizontally in an S shape during distant detection. Please refer to Figure 5 as shown in Figure 5 is a schematic diagram of calibrating the driving trajectory of the detection target onto the actual road of the high-precision map. Although the calibration is accurate macroscopically at this time, within each lane, the radar data still cannot be accurately mapped to the specific lane line, and lane-level accuracy calibration cannot be achieved. Therefore, based on Embodiment 1, Embodiment 2 provides a method for calibrating a radar based on a high-precision map. Please refer to Figure 4 as shown in Figure 4 is a schematic flowchart of a method for calibrating a radar based on a high-precision map provided by Embodiment 2, which realizes lane-level calibration by correcting the detection target according to the high-precision map data. The method includes the following steps:

[0061] S201. Calibrate the angle of the radar using the third angle.

[0062] Specifically, it is basically the same as the technical solution in Embodiment 1. Please refer to the description of the relevant content in Embodiment 1, and it will not be repeated here.

[0063] S203. Obtain the second historical trajectory of the clustered detection target after angle calibration.

[0064] For the specific acquisition method, please refer to Embodiment 1, and it will not be repeated here.

[0065] S205. Determine whether the second historical trajectory jitters. If so, perform secondary calibration on the radar by aligning the second historical trajectory with jitter to the center of the lane line of the actual road.

[0066] Specifically, the detection of distant objects by the radar may show lateral jitter. When it is found in the detection that the historical trajectory of the detection target jitters, the lane line information in the high-precision map is used as a reference, and the detection result of the jittering radar is forced to be aligned to the center of the lane line. The target heading angle information is mainly based on the tangent direction of the lane line. The calibration of the long-distance detection of the radar is completed.

[0067] Further, during the process of aligning to the lane line, when the detection target presses the line, the detection target is also aligned to the center of the lane line.

[0068] In the embodiment of the present disclosure, by using the high-precision map as a reference, the detection target at a long distance of the radar is forced to be aligned to the center of the lane line, the radar is calibrated, the purpose of filtering jitter and heading angle deviation is achieved, the purpose of improving the radar detection accuracy is achieved, and the calibration with the detection accuracy accurate to within the lane line is realized.

[0069] It should also be noted that if the detection target does not jitter during radar detection, the radar is in a normal state and no operation is performed.

[0070] It should be understood that the above-listed relevant specific contents are only for illustrative purposes and should not impose any limitation on the present invention.

[0071] Embodiment 3

[0072] During the cluster deployment of the radar, it is inevitable that the detection areas of the radars overlap, which will cause deviations in the detection results of the detection target in the overlapping area of two adjacent radars, that is, the two detection results do not overlap. Since this area is generally hundreds of meters away from one of the radars. Traditional calibration schemes are difficult to implement. Therefore, on the basis of Embodiment 2, Embodiment 3 provides a method for calibrating the radar based on a high-precision map to achieve the purpose of matching the detection results of the detection target in the overlapping detection areas of two adjacent radars during the multi-pole relay of the radar. Please refer to Figure 6 as shown in Figure 6 is a schematic flowchart of a method for calibrating a radar based on a high-precision map provided in Embodiment 3 of this specification; the method includes the following steps:

[0073] S301. Obtain the first position data detected by the first radar after secondary calibration for the clustering detection target;

[0074] Specifically, on the basis of meeting the lane line calibration accuracy, the first position data detected by the first radar for the detection target is obtained, and the first position data is projected onto the actual lane of the high-precision map.

[0075] S303. Obtain the second position data detected by the second radar after secondary calibration for the clustering detection target, where the first radar and the second radar are adjacent;

[0076] Specifically, since the detection target is within the overlapping area of the first radar and the second radar, a second detection result, that is, the second position data, will appear. The second position data is projected onto the actual lane of the high-precision map.

[0077] S305. Obtain a position error based on the first position data and the second position data;

[0078] Specifically, the first position data and the second position data are subtracted to obtain the position error. Due to the constraint of the lane line, there will be no lateral error between the first position data and the second position data. The error occurs in the forward direction. If there is an error, it will cause the two detection results not to overlap.

[0079] S307. Determine whether the position error meets a preset position error. If not, perform distance gain adjustment and time offset adjustment on the first radar or / and the second radar, and perform three calibrations on the first radar or / and the second radar to make the position error meet the preset position error.

[0080] Specifically, the preset position error is half of the vehicle body. If the distance exceeds half of the vehicle body, the detection results of the two radars do not meet the requirements. Then, perform distance gain adjustment and time offset adjustment on the radar to make the position error meet the preset position error.

[0081] Further, within the overlapping area of the first radar and the second radar, after the adjacent two radars continuously detect the detection target for a period of time, such as a continuous detection time of 1 s (about 10 frames). At this time, if the position error does not meet the requirements and the distance between the detection target and the first radar is less than the distance between the detection target and the second radar, then the distance gain adjustment and time gain adjustment are performed on the second radar based on the detection result of the first radar to make the position error meet the preset position error.

[0082] To further illustrate, within the overlapping area of the first radar and the second radar, after two adjacent radars continuously detect a detection target for a period of time, such as a continuous detection time of 1 s (about 10 frames), at this time, if the position error does not meet the requirements, and the distance between the detection target and the first radar is greater than the distance between the detection target and the second radar, then the distance gain adjustment and time gain adjustment are performed on the first radar based on the detection result of the second radar, so that the position error meets the preset position error.

[0083] Among them, the distance gain adjustment can be understood as that if the overlapping perception areas of the two radars are quite different, the longitudinal distance is the distance along the lane line direction multiplied by a coefficient, so that the position error meets the preset position error. The time gain can be understood as that if it is found that the deviation of a high-speed vehicle is large and the deviation of a low-speed vehicle is small, the target timestamp is advanced or postponed, so that the position error meets the preset position error.

[0084] Preferably, the detection bandwidths of adjacent radars should be staggered. The radars are connected to the same time synchronization server. Perform NTP time synchronization with a synchronization error within the millimeter level. After the radar goes through a sampling period and the high-speed ADC samples the reflected wave, a timestamp is marked before the FFT calculation. This timestamp always follows the detection target of this sampling to improve the detection accuracy.

[0085] In the embodiment of the present disclosure, based on the high-precision map, the detection results of the overlapping areas of two adjacent radars are projected onto the high-precision map. If the error is large, distance gain adjustment and time offset adjustment are performed so that the position error meets the preset position error; the purpose of target matching in the overlapping perception areas of multiple radars on the road network is achieved.

[0086] It should be understood that the above-listed relevant specific contents are only for illustrative purposes and should not impose any limitation on the present invention.

[0087] Embodiment 4

[0088] Since the radar is installed outdoors, after being exposed to wind, rain and sunlight for a long time, the angle will deviate from the initial installation angle, resulting in inaccurate detection results. Therefore, on the basis of Embodiment 2, Embodiment 4 provides a method for calibrating the radar based on a high-precision map to solve the problem that the initial installation angle deviates due to long-term exposure to wind, rain and sunlight. Please refer to Figure 7 as shown in Figure 7 is a schematic flowchart of a method for calibrating a radar based on a high-precision map provided in Embodiment 4 of this specification; the method includes the following steps:

[0089] S401. Construct the initial quadratic function of the calibrated radar, and obtain the initial inclination rate through the initial quadratic function;

[0090] Specifically, a rectangular coordinate system is established with the radar after secondary calibration as the origin; the third historical trajectory of the clustering detection target is obtained, where the third historical trajectory refers to the trajectory of the radar after secondary calibration detecting the clustering detection target; an initial least squares function is constructed on the rectangular coordinate system through the third historical trajectory. Further, after the first installation and calibration, with the installation position of the radar after secondary calibration as the coordinate origin, the due east direction as the X-axis, and the due north direction as the Y-axis, the least squares method is applied to the third historical trajectory of the radar detecting an object. Here, the least squares method is the initial least squares function, and the initial least squares function is the same as the corresponding least squares function of the lane line. The initial least squares function is obtained as Y = k*X + b, and the k and b of the function are recorded.

[0091] S403. Construct the least squares function to be measured of the radar after secondary calibration, and obtain the inclination rate to be measured through the least squares function to be measured;

[0092] Specifically, after the radar angle has a slight deviation after a period of use, the least squares method is applied to the trajectory of the radar detecting an object, and the least squares function to be measured is obtained as Y = k1*X + b1.

[0093] S405. Obtain the deviation of the tilt angle based on the inclination rate to be measured and the initial inclination rate;

[0094] Specifically, the inclination rate to be measured is compared with the initial inclination rate to obtain the deviation of the tilt angle as Figure 8 shown, Figure 8 This is the projection schematic diagram of the least squares function of a method for calibrating a radar based on a high-precision map provided in Embodiment 4 of this specification. The least squares function to be measured and the initial least squares function are projected onto the high-precision map, and then the least squares function to be measured with an inclination rate of K1 is compared with the initial least squares function with an inclination rate of K to obtain the deviation of the two tilt angles.

[0095] S407. Calibrate the position of the radar using the deviation of the tilt angle.

[0096] Specifically, the deviation of the tilt angle is input into the radar to make K1 equal to K, and the position of the radar is calibrated.

[0097] In the embodiments of the present disclosure, based on the high-precision map, the parameters of the least squares equation of the trajectory of the radar detecting an object are obtained. Then, it is calculated once every other period of time. According to this value and the initial deviation, the slight offset of the radar caused by long-term outdoor use is automatically calibrated. The labor cost is reduced, and the detection accuracy of the radar is improved.

[0098] It should be understood that the above-listed relevant specific contents are only for illustrative purposes and should not impose any limitations on the present invention.

[0099] Embodiment 5

[0100] Embodiment 5 provides a device for calibrating a roadside millimeter-wave radar based on a high-precision map. Please refer to Figure 9 as shown in Figure 9 FIG. Figure 9 is a schematic structural diagram of a device for calibrating a roadside millimeter-wave radar based on a high-precision map provided in Embodiment 5. The device includes:

[0101] A trajectory processing module 501, configured to obtain a first historical trajectory of a clustering detection target, and form a topological map according to the first historical trajectory of the clustering detection target, where the clustering detection target refers to a target within the detection range of the radar;

[0102] A comparison module 503, configured to compare the topological map with an actual road in the high-precision map to obtain a first included angle between the topological map and the actual road, where the actual road refers to the road corresponding to the first historical trajectory;

[0103] An extraction module 505, configured to obtain a second included angle between the actual road and a preset direction;

[0104] A calculation module 507, configured to obtain a third included angle between the historical trajectory of the detection target and the preset direction according to the first included angle and the second included angle;

[0105] A calibration module 509, configured to perform angle calibration on the radar by using the third included angle.

[0106] A judgment module, configured to judge whether jitter occurs when the second historical trajectory is present.

[0107] The extraction module is further configured to obtain first position data detected by the first radar after secondary calibration for the clustering detection target;

[0108] Obtain second position data detected by the second radar after secondary calibration for the clustering detection target, where the first radar and the second radar are adjacent.

[0109] The comparison module is further configured to obtain a position error according to the first position data and the second position data.

[0110] The judgment module is further configured to judge whether the position error meets a preset position error. If not, perform distance gain adjustment and time offset adjustment on the first radar or / and the second radar.

[0111] A model construction module, configured to construct an initial least-squares function of the radar after secondary calibration, and obtain an initial inclination rate through the initial least-squares function;

[0112] Construct a to-be-detected least-squares function of the radar after secondary calibration, and obtain a to-be-detected inclination rate through the to-be-detected least-squares function;

[0113] The comparison module is further configured to obtain an inclination angle deviation based on the inclination rate to be measured and the initial inclination rate.

[0114] Embodiment 6

[0115] Embodiment 6 provides an electronic device, including at least one processor and a memory. The memory stores a program and is configured to enable the at least one processor to execute a method for calibrating a radar based on a high-precision map according to any one of the foregoing embodiments.

[0116] Embodiment 7

[0117] Embodiment 7 provides a computer-readable storage medium storing computer instructions for causing a computer to execute a method for calibrating a radar based on a high-precision map according to any one of the foregoing embodiments.

[0118] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to circuit structures such as diodes, transistors, switches, etc.) or software improvements (improvements to method flows). However, with the development of technology, many method flow improvements today can be regarded as direct improvements to hardware circuit structures. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement to a method flow cannot be implemented using a hardware entity module. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is an integrated circuit whose logic function is determined by the user programming the device. Designers can program themselves to "integrate" a digital system onto a single PLD, without having to ask a chip manufacturer to design and fabricate a dedicated integrated circuit chip. Moreover, nowadays, instead of manually fabricating integrated circuit chips, this programming is mostly implemented using "logic compiler" software, which is similar to the software compiler used in program development and writing. The original code before compilation also has to be written in a specific programming language, which is called a Hardware Description Language (HDL), and there is not just one type of HDL, but many, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones currently are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also be aware that by simply making a little logical programming of the method flow using the above-mentioned several hardware description languages and programming it into the integrated circuit, it is easy to obtain the hardware circuit that implements the logical method flow.

[0119] The controller can be implemented in any suitable manner. For example, the controller can take the form of, for example, a microprocessor or a processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of the controller include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that, in addition to implementing the controller in the form of pure computer-readable program code, it is entirely possible to implement the same function by logically programming the method steps so that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be regarded as the structures within the hardware component. Or even, the devices for implementing various functions can be regarded as either software modules for implementing the method or structures within the hardware component.

[0120] The systems, devices, modules, or units illustrated in the above description of the embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. Among them, a typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0121] For the convenience of description, when describing the above devices, they are divided into various modules or units according to functions and described separately. Of course, when implementing the present application, the functions of each module or each unit can be implemented in the same or multiple software and / or hardware.

[0122] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0123] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present specification. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, as well as the combination of flows and / or blocks in the flowchart and / or block diagram. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0124] These computer program instructions can 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, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented processing flow, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0126] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0127] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory (NVM), such as read-only memory (ROM) or flash RAM. The memory is an example of computer-readable media.

[0128] A computer-readable medium includes both permanent and non-permanent, removable and non-removable media and can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information accessible by a computing device. As defined herein, a computer-readable medium does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0129] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article or apparatus comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or apparatus. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising the element.

[0130] This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This application can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0131] Each embodiment in this specification is described in a progressive manner, and the same or similar parts among the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and the relevant parts can refer to the description of the method embodiments.

[0132] The above description is only for the embodiments of the specification of the present application and is not intended to limit the present application. For those skilled in the art, various modifications and variations can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for calibrating a radar based on a high-precision map, characterized in that, The method includes the following steps: Obtain the first historical trajectory of the clustering detection target, and form a topological map according to the first historical trajectory of the clustering detection target, where the clustering detection target refers to the target within the radar detection range; Compare the topological map with the actual road in the high-precision map to obtain the first included angle between the topological map and the actual road, where the actual road refers to the road corresponding to the first historical trajectory; Obtain the second included angle between the actual road and the preset direction; Obtain the third included angle between the historical trajectory of the detection target and the preset direction according to the first included angle and the second included angle; Calibrate the angle of the radar using the third included angle; Obtain the second historical trajectory of the clustering detection target after angle calibration; Judge whether the second historical trajectory jitters. If so, center the jittered second historical trajectory on the center line of the lane of the actual road, and perform secondary calibration on the radar; Obtain the first position data detected by the first radar after secondary calibration for the clustering detection target; Obtain the second position data detected by the second radar after secondary calibration for the clustering detection target, where the first radar and the second radar are adjacent; Obtain the position error according to the first position data and the second position data; Judge whether the position error meets the preset position error. If not, perform distance gain adjustment and time offset adjustment on the first radar or / and the second radar, and perform three calibrations on the first radar or / and the second radar.

2. The method for calibrating a radar based on a high-precision map according to claim 1, wherein After performing the secondary calibration on the radar, it further includes: Construct an initial least-squares function of the radar after secondary calibration, and obtain an initial inclination rate through the initial least-squares function; Construct a to-be-detected least-squares function of the radar after secondary calibration, and obtain a to-be-detected inclination rate through the to-be-detected least-squares function; Obtain the inclination angle deviation according to the to-be-detected inclination rate and the initial inclination rate; Perform position calibration on the radar using the inclination angle deviation.

3. The method for calibrating a radar based on a high-precision map according to claim 2, wherein, The construction of the initial least-squares function of the radar after secondary calibration includes: Establish a rectangular coordinate system with the radar after secondary calibration as the origin; Obtain the third historical trajectory of the clustering detection target, where the third historical trajectory refers to the trajectory detected by the radar after secondary calibration for the clustering detection target; Construct an initial least-squares function on the rectangular coordinate system through the third historical trajectory.

4. A method for calibrating a radar based on a high-precision map according to claim 1, characterized in that, Before obtaining the first historical trajectory of the clustering detection target, it further includes: clustering multiple distributed detection targets to form a clustering detection target.

5. A method for calibrating a radar based on a high-precision map according to claim 4, characterized in that, The clustering of multiple distributed detection targets includes: obtaining the distance between two adjacent distributed detection targets, and judging whether the distance exceeds the preset length. If not, cluster the two adjacent distributed detection targets into the same type of target.

6. An apparatus for calibrating a radar based on a high-precision map, characterized in that, It includes: A trajectory processing module, configured to obtain the first historical trajectory of the clustering detection target, and form a topological map according to the first historical trajectory of the clustering detection target, where the clustering detection target refers to the target within the radar detection range; A comparison module, configured to compare the topological map with the actual road in the high-precision map to obtain the first included angle between the topological map and the actual road, where the actual road refers to the road corresponding to the first historical trajectory; An extraction module for obtaining a second included angle between the actual road and a preset direction; A calculation module for obtaining a third included angle between the historical trajectory of the detection target and the preset direction according to the first included angle and the second included angle; A calibration module for performing angle calibration on the radar by using the third included angle; The calibration module is further configured to: Obtain a second historical trajectory of the clustered detection target after angle calibration; Determine whether the second historical trajectory jitters. If so, perform secondary calibration on the radar by aligning the jittered second historical trajectory with the center of the lane line of the actual road; Obtain first position data detected by the first radar on the clustered detection target after secondary calibration; Obtain second position data detected by the second radar on the clustered detection target after secondary calibration, where the first radar and the second radar are adjacent; Obtain a position error according to the first position data and the second position data; Determine whether the position error meets a preset position error. If not, perform distance gain adjustment and time offset adjustment on the first radar or / and the second radar, and perform tertiary calibration on the first radar or / and the second radar.

7. An electronic device, characterized in that, Comprising at least one processor and a memory, the memory stores a program and is configured such that at least one processor is used to cause a computer to execute a method for calibrating a radar based on a high-precision map according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute a method for calibrating a radar based on a high-precision map according to any one of claims 1-5.

Images