Method for dynamic calibration of millimeter wave radar, radar system, vehicle and computer program product

Abstract

The invention relates to a method for the dynamic calibration of a millimeter-wave radar, comprising the following steps: screening out static point cloud data from each frame of radar point cloud data acquired by the millimeter-wave radar (11) under a predetermined driving condition of a vehicle (10), and determining a proportion parameter of each static point cloud data, the proportional parameter is used for representing the ratio (S1) of the Doppler velocity vd of the static point cloud data to the velocity vh of the vehicle; screening out valid static point cloud data on the basis of the obtained proportional parameter (S2); and ascertaining the installation deviation of the millimeter-wave radar by means of a Taylor series expansion of a cosine function on the basis of the proportional parameter of the valid static point cloud data (S3). According to the method, dynamic calibration of the millimeter-wave radar can be efficiently realized by means of the Taylor series expansion of the cosine function, the horizontal angle installation deviation and / or the pitch angle installation deviation of the millimeter-wave radar can be continuously and effectively calibrated, and particularly, a large amount of complex matrix operation is omitted.

Description

Technical Field

[0001] The present application relates to the field of vehicle-mounted radar, and in particular to a method for dynamic calibration of a millimeter-wave radar, a radar system, a vehicle including the radar system, and a computer program product for at least assisting in implementing the steps of the method described in the present application. Background Art

[0002] During the installation of vehicle-mounted radar, the installation deviation of the vehicle-mounted radar is usually caused by the limitation of installation technology and the tolerance of vehicle size, that is, the normal of the detection surface of the vehicle-mounted radar is not completely along the front direction of the vehicle. This installation angle deviation will directly affect the angle information of all targets measured by the vehicle-mounted radar, resulting in an overall offset, and this offset will increase with the increase of the distance of the target, thereby affecting the normal execution of related functions of the advanced driver assistance system. Existing radar calibration methods are usually more complicated and have strict requirements on the calibration site.

[0003] Therefore, there is room for improvement in the current radar calibration method. Summary of the invention

[0004] The object of the present application is to provide a method for dynamic calibration of a millimeter-wave radar, a radar system, a vehicle including the radar system, and a computer program product, so as to at least partially solve the problems in the prior art.

[0005] According to a first aspect of the present application, a method for dynamic calibration of a millimeter wave radar is provided, the method comprising:

[0006] - Step S1: Under a predetermined driving condition of the vehicle, static point cloud data are selected from each frame of radar point cloud data acquired by the millimeter wave radar, and a proportional parameter of each static point cloud data is obtained, wherein the proportional parameter is used to characterize the Doppler velocity v of the static point cloud data d With the vehicle's speed v h The ratio of

[0007] - Step S2: Filter out valid static point cloud data based on the obtained scale parameter; and

[0008] - Step S3: Calculating the installation deviation of the millimeter wave radar based on the scale parameter of the valid static point cloud data by means of the Taylor series expansion of the cosine function.

[0009] The core concept of this application is: based on the Doppler information and angle information provided by the screened static point cloud data, the Taylor series expansion of the cosine function can efficiently realize the dynamic calibration of the millimeter wave radar, and continuously and effectively calibrate the horizontal angle installation deviation and / or the pitch angle installation deviation of the millimeter wave radar. Compared with the traditional cosine matching algorithm, the calibration method according to this application has a relatively simple calculation process, especially eliminating a large number of complex matrix operations required in the cosine matching algorithm to accumulate cosine curves for parabola fitting.

[0010] According to an embodiment of the present application, step S2 may include:

[0011] - Step S21: sorting the static point cloud data in each frame of radar point cloud data based on the obtained proportional parameter, and determining the maximum proportional parameter assigned to each frame;

[0012] - Step S22: Comparing the maximum proportional variable assigned to each frame with a predefined cosine threshold value, wherein the predefined cosine threshold value is, for example, cos6°; and

[0013] Step S23 : when the maximum proportional variable associated with a frame is greater than or equal to a predefined cosine threshold value, the static point cloud data having the maximum proportional variable are determined as valid static point cloud data associated with the frame.

[0014] According to another embodiment of the present application, the installation deviation Δθ of the millimeter wave radar assigned to each frame can be obtained by using the Taylor series expansion of the cosine function based on the valid static point cloud data assigned to each frame. i , and calculate the installation deviation Δθ of the millimeter wave radar assigned to each frame i The average value is taken as the installation deviation Δθ of the millimeter-wave radar.

[0015] According to another embodiment of the present application, each frame of radar point cloud data acquired by the millimeter-wave radar is respectively distributed in each angle interval of the horizontal field of view angle and / or the vertical field of view angle of the millimeter-wave radar, wherein each angle interval occupies an angle between 1° and 10°, in particular 1°.

[0016] According to another embodiment of the present application, when the valid static point cloud data is radar point cloud data obtained in the horizontal field of view angle of the millimeter-wave radar, the azimuth installation deviation of the millimeter-wave radar can be calculated based on the scale parameters of the valid static point cloud data with the help of the Taylor series expansion of the cosine function; when the valid static point cloud data is radar point cloud data obtained in the vertical field of view angle of the millimeter-wave radar, the pitch installation deviation of the millimeter-wave radar can be calculated based on the scale parameters of the valid static point cloud data with the help of the Taylor series expansion of the cosine function.

[0017] According to another embodiment of the present application, the azimuth installation deviation and / or the elevation installation deviation Δθ of the millimeter wave radar can be obtained based on the scale parameter of the effective static point cloud data by means of a quadratic function approximation of the Taylor series expansion of the cosine function, wherein the obtaining process is described as the following equation:

[0018]

[0019] Wherein, θ′ represents the angle value of the angle interval in which the valid static point cloud data is located in the horizontal field of view angle and / or the vertical field of view angle of the millimeter wave radar.

[0020] According to another embodiment of the present application, the predetermined driving condition of the vehicle includes: the speed v of the vehicle h within a predetermined speed range, the predetermined speed range is, for example, 15 km / h to 120 km / h; and / or the yaw angle change rate of the vehicle is within a predetermined yaw angle change rate range, the predetermined yaw angle change rate range is, for example, less than 0.03 rad / s; and / or the acceleration of the vehicle is within a predetermined acceleration range, the predetermined yaw angle change rate range is, for example, less than ±5 m / s 2 ,etc.

[0021] According to a second aspect of the present application, a radar system is provided, wherein the radar system includes a millimeter wave radar and a controller for executing the method according to the present application.

[0022] According to a third aspect of the present application, a vehicle is provided, comprising a radar system according to the present application.

[0023] According to a fourth aspect of the present application, a computer program product, such as a computer-readable program carrier, is provided, which contains or stores computer program instructions, and when the computer program instructions are executed by a processor, they at least assist in implementing the steps of the method according to the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The principles, features and advantages of the present application may be better understood by describing the present application in more detail below with reference to the accompanying drawings. The accompanying drawings show:

[0025] Figure 1 A flowchart showing a method for dynamic calibration of a millimeter wave radar according to an exemplary embodiment of the present application is shown;

[0026] Figure 2 A distribution diagram of a radar point cloud acquired by a millimeter wave radar according to an exemplary embodiment of the present application is shown;

[0027] Figure 3 A schematic diagram of a calibration scenario according to an exemplary embodiment of the present application is shown;

[0028] Figure 4 A flowchart showing a method for dynamic calibration of a millimeter wave radar according to another exemplary embodiment of the present application is shown;

[0029] Figure 5 An angle interval distribution diagram showing the field of view angle of a millimeter wave radar according to an exemplary embodiment of the present application;

[0030] Figure 6 A schematic block diagram showing a radar system according to an exemplary embodiment of the present application; and

[0031] Figure 7 A schematic diagram of a vehicle according to an exemplary embodiment of the present application is shown. DETAILED DESCRIPTION

[0032] In order to make the technical problems, technical solutions and beneficial technical effects to be solved by the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and multiple exemplary embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the scope of protection of the present application.

[0033] Figure 1 A flowchart of a method for dynamic calibration of a millimeter wave radar according to an exemplary embodiment of the present application is shown. The following exemplary embodiments describe the method according to the present application in more detail.

[0034] In step S1, under a predetermined driving condition of the vehicle 10, static point cloud data are screened out from each frame of radar point cloud data acquired by the millimeter wave radar 11, and a proportional parameter of each static point cloud data is obtained, wherein the proportional parameter is used to characterize the Doppler velocity v of the static point cloud data. d With the speed v of the vehicle 10 h ratio.

[0035] The calibration method according to the present application is only applicable to millimeter wave radars because the calibration method utilizes the Doppler velocity v of the radar point cloud data. d Divide by the speed v of the vehicle 10 h The basic principle is equal to the cosine value of the azimuth angle α and / or elevation angle β of the radar point cloud data, that is, the following formula:

[0036] -v d / v h = cosα or -v d / v h =cosβ,

[0037] Other types of radar sensors (such as lidar sensors, ultrasonic radars, etc.) cannot detect the Doppler velocity of the target object, and therefore cannot use this basic principle.

[0038] Before executing the calibration method according to the present application, it is necessary to monitor the operating conditions of the vehicle 10. The calibration method according to the present application can be executed only when the operating conditions of the vehicle 10 meet the predetermined operating conditions, especially when the vehicle 10 is moving forward steadily in a straight line. For example, the predetermined driving conditions of the vehicle 10 may include: the speed v of the vehicle 10 h within a predetermined speed range, the predetermined speed range is, for example, 15 km / h to 120 km / h; and / or the yaw angle change rate of the vehicle 10 is within a predetermined yaw angle change rate range, the predetermined yaw angle change rate range is, for example, less than 0.03 rad / s; and / or the acceleration of the vehicle 10 is within a predetermined acceleration range, the predetermined yaw angle change rate range is, for example, less than ±5 m / s 2 ,etc.

[0039] When the vehicle 10 is detected to be in a predetermined driving condition, static point cloud data is screened out from each frame of radar point cloud data acquired by the millimeter wave radar 11. In this application, the radar point cloud data acquired by the millimeter wave radar 11 about a stationary target is static point cloud data, and the radar point cloud data acquired about a moving target is dynamic point cloud data. Here, the Doppler velocity v of each frame of radar point cloud data can be obtained. d With the speed v of the vehicle 10 h and filter out the static point cloud data based on the ratio and the azimuth angle α and / or elevation angle β of the radar point cloud data. Figure 2 As shown, the ratio -v d / v h The radar point cloud distribution diagram is drawn with the azimuth angle α of the radar point cloud data as the vertical axis and, for example, the radar point cloud data between two preset boundary curves (marked with hollow circles) as static point cloud data about static targets, and the radar point cloud data outside the two preset boundary lines (marked with solid circles) as dynamic point cloud data about dynamic targets, wherein there are multiple disclosed implementation schemes for setting the two preset boundary lines, which will not be described here. The static point cloud data is also filtered in the same way for the pitch angle β, which will not be described here again.

[0040] After filtering out the static point cloud data, the proportional parameters of each static point cloud data can be obtained, which are used to characterize the Doppler velocity v of the static point cloud data. d With the speed v of the vehicle 10 h The ratio of, for example, -vd / v h .

[0041] In step S2, valid static point cloud data can be screened out based on the obtained scale parameter. Figure 3 FIG. 1 is a schematic diagram of a calibration scene according to an exemplary embodiment of the present application, in which the forward millimeter-wave radar 11 of the vehicle 10 is calibrated, wherein the x-axis direction is the driving direction of the vehicle 10, and the hollow circle is static point cloud data, which has a Doppler velocity v of the forward millimeter-wave radar 11 toward the vehicle 10. d , the Doppler velocity v d The direction and speed v of the vehicle 10 h The real angle in the direction is marked as θ. Considering that the installation angle deviation of the millimeter-wave radar is usually within the angle range of ±6°, when the real angle θ between the Doppler velocity direction of the static point cloud data and the vehicle's driving direction is within the angle range of ±6°, it can be considered that the real angle is small enough, thereby screening out valid static point cloud data.

[0042] The following combination Figure 4 A workflow diagram showing a method for dynamic calibration of a millimeter-wave radar according to another exemplary embodiment of the present application elaborates a screening process of valid static point cloud data.

[0043] like Figure 4 As shown, step S2 may include steps S21 to S23. In step S21, the static point cloud data in each frame of radar point cloud data is sorted based on the obtained scale parameter, and the maximum scale parameter assigned to each frame is determined. Here, for each frame of radar point cloud data, only one static point cloud data with the maximum scale parameter is selected to evaluate the validity of the frame of radar point cloud data.

[0044] In step S22 , the maximum proportional variable associated with each frame is compared with a predefined cosine threshold value, wherein the predefined cosine threshold value is, for example, cos6°.

[0045] When the maximum proportional parameter assigned to a frame is greater than or equal to a predetermined cosine threshold, this means that the actual angle θ between the Doppler velocity direction of the static point cloud data with the maximum proportional parameter and the vehicle's driving direction is within the angular range of ±6°. Therefore, it can be determined that this frame of radar point cloud data is valid, and in step S23, the static point cloud data with the maximum proportional parameter is determined as the valid static point cloud data assigned to the frame.

[0046] When the maximum proportional parameter assigned to a frame is less than a predetermined cosine threshold, this means that the actual angle θ between the Doppler velocity direction of the static point cloud data with the maximum proportional parameter and the vehicle's driving direction exceeds the angle range of ±6°. Therefore, it can be determined that the frame of radar point cloud data is invalid, and returns to step S1 to wait for the next frame of radar point cloud data to be processed.

[0047] In step S3, the installation deviation of the millimeter wave radar 11 can be obtained based on the scale parameter of the effective static point cloud data by using the Taylor series expansion of the cosine function. According to the basic principle of millimeter wave radar, the Doppler velocity v of the effective static point cloud data is d Divide by the speed v of the vehicle 10 h is equal to the cosine value of the true angle θ between the Doppler velocity direction of the valid static point cloud data and the vehicle's driving direction, that is,

[0048] -v d / v h = cosθ

[0049] like Figure 5 The angle interval distribution diagram of the field of view of the millimeter wave radar according to an exemplary embodiment of the present application is shown. In the calibration scenario of the azimuth angle α of the millimeter wave radar 11, the horizontal field of view angle of the millimeter wave radar 11 can be divided into multiple angle intervals, wherein each angle interval can occupy an angle between 1° and 10°, especially 1°. In the case where the valid static point cloud data is the radar point cloud data obtained in the horizontal field of view angle of the millimeter wave radar 11, the real angle θ between the Doppler velocity direction of the valid static point cloud data and the driving direction of the vehicle can be considered as the sum of the angle value θ′ (which is the azimuth detected by the millimeter wave radar 11) of the angle interval where the valid static point cloud data is located in the horizontal field of view angle of the millimeter wave radar 11 and the azimuth installation deviation Δθ of the millimeter wave radar 11, thereby the azimuth installation deviation Δθ of the millimeter wave radar 11 can be obtained based on the scale parameter of the valid static point cloud data with the help of the Taylor series expansion of the cosine function, wherein the Taylor series expansion of the cosine function of the real angle θ is:

[0050]

[0051] In order to simplify the calculation, the stationary target detected by the millimeter wave radar 11 should be as far as possible in front of the vehicle (i.e. Figure 3), and the heights of the various stationary targets are roughly equal, that is, they are roughly in the same horizontal plane, so it can be assumed that the true angle θ between the Doppler velocity direction of the effective static point cloud data and the vehicle travel direction in the horizontal plane is close to 0°. Therefore, the azimuth installation deviation Δθ of the millimeter-wave radar 11 can be obtained based on the scale parameters of the effective static point cloud data by means of the quadratic function approximation of the Taylor series expansion of the cosine function, wherein the obtaining process is described as the following equation:

[0052]

[0053] For example, if the valid static point cloud data is selected from the i-th frame of radar point cloud data, the azimuth installation deviation Δθ obtained is the azimuth installation deviation Δθ assigned to the i-th frame. i The azimuth installation deviation Δθ of the millimeter wave radar 11 assigned to each frame can be obtained. i The average value is taken as the azimuth installation deviation Δθ of the millimeter wave radar 11.

[0054] Similarly, in the calibration scenario of the pitch angle β of the millimeter-wave radar 11, the vertical field of view angle of the millimeter-wave radar 11 can also be divided into multiple angle intervals, wherein each angle interval can occupy an angle between 1° and 10°, especially 1°. In the case where the valid static point cloud data is the radar point cloud data obtained in the vertical field of view angle of the millimeter-wave radar 11, the true angle θ between the Doppler velocity direction of the valid static point cloud data and the vehicle driving direction can be considered to be the sum of the angle value θ′ (which is the pitch angle detected by the millimeter-wave radar 11) of the angle interval where the valid static point cloud data is located in the vertical field of view angle of the millimeter-wave radar 11 and the pitch angle installation deviation Δθ of the millimeter-wave radar 11, thereby the pitch angle installation deviation Δθ of the millimeter-wave radar 11 can be obtained based on the scale parameter of the valid static point cloud data with the help of the Taylor series expansion of the cosine function.

[0055] In order to simplify the calculation, the stationary targets detected by the millimeter wave radar 11 should be in the same vertical plane parallel to the forward direction of the vehicle 1 as much as possible, and the height of each stationary target is roughly equal to the installation height of the millimeter wave radar 11, so that it can be assumed that the real angle θ between the Doppler velocity direction of the effective static point cloud data and the vehicle's driving direction in the vertical plane is close to 0°. Therefore, the pitch angle installation deviation Δθ of the millimeter wave radar 11 can be obtained based on the scale parameter of the effective static point cloud data with the help of the quadratic function approximation of the Taylor series expansion of the cosine function, wherein the obtaining process is also described as the following equation:

[0056]

[0057] For example, if the valid static point cloud data is selected from the i-th frame of radar point cloud data, the pitch angle installation deviation Δθ obtained is the pitch angle installation deviation Δθ assigned to the i-th frame. i The elevation angle installation deviation Δθ of the millimeter wave radar 11 assigned to each frame can be obtained. i The average value is taken as the elevation angle installation deviation Δθ of the millimeter wave radar 11.

[0058] According to the embodiments of the present application, based on the Doppler information and angle information provided by the filtered static point cloud data, the dynamic calibration of the millimeter wave radar can be efficiently realized by using the Taylor series expansion of the cosine function, and the horizontal angle installation deviation and / or the pitch angle installation deviation of the millimeter wave radar can be calibrated continuously and effectively. Compared with the traditional cosine matching algorithm, the calibration method according to the present application has a relatively simple operation process, especially eliminating a large number of complex matrix operations required in the cosine matching algorithm to accumulate cosine curves for parabola fitting.

[0059] In addition, it should be noted that the step numbers described herein do not necessarily represent a chronological order, but are merely a figure mark. The order can be changed according to specific circumstances as long as the technical purpose of the present application can be achieved.

[0060] Figure 6 A schematic block diagram of a radar system 1 according to an exemplary embodiment of the present application is shown.

[0061] like Figure 6 As shown, the radar system 1 may include a millimeter wave radar 11 and a controller 12 for executing the method according to the present application.

[0062] Figure 7 A schematic diagram of a vehicle according to an exemplary embodiment of the present application is shown.

[0063] like Figure 7 As shown, the vehicle 10 may include the radar system 1 according to the present application.

[0064] If an embodiment includes an “and / or” relationship between a first feature and a second feature, it should be interpreted as follows: according to one embodiment, the embodiment has both the first feature and the second feature, and according to another embodiment, the embodiment has either only the first feature or only the second feature.

[0065] Although specific embodiments have been described above, these embodiments are not intended to limit the scope of the present disclosure, even when only a single embodiment is described with respect to specific features. The feature examples provided in the present disclosure are intended to be illustrative, not restrictive, unless otherwise stated. In specific implementations, multiple features may be combined with each other, where technically feasible, according to actual needs. Various substitutions, changes, and modifications may also be conceived without departing from the spirit and scope of the present application.

Claims

1. A method for dynamic calibration of a millimeter wave radar (11), the method comprising: Step S1: Under a predetermined driving condition of the vehicle (10), static point cloud data are selected from each frame of radar point cloud data acquired by the millimeter wave radar (11), and a proportional parameter of each static point cloud data is obtained, wherein the proportional parameter is used to characterize the Doppler velocity v of the static point cloud data. d With the speed v of the vehicle (10) h The ratio of Step S2: Filter out valid static point cloud data based on the obtained scale parameter; and Step S3: Based on the scale parameter of the effective static point cloud data, the installation deviation of the millimeter wave radar (11) is calculated by using the Taylor series expansion of the cosine function.

2. The method according to claim 1, wherein: The step S2 comprises: Step S21: sorting the static point cloud data in each frame of radar point cloud data based on the obtained proportional parameter, and determining the maximum proportional parameter assigned to each frame; Step S22: Compare the maximum proportional variable assigned to each frame with a predetermined cosine threshold value, wherein the predetermined cosine threshold value is, for example, cos6°; and Step S23 : when the maximum proportional parameter associated with a frame is greater than or equal to a predefined cosine threshold value, determining the static point cloud data having the maximum proportional parameter as valid static point cloud data associated with the frame.

3. A method according to any one of the preceding claims, wherein: Based on the effective static point cloud data assigned to each frame, the installation deviation Δθ of the millimeter wave radar (11) assigned to each frame is calculated by using the Taylor series expansion of the cosine function. i , and obtain the installation deviation Δθ of the millimeter wave radar (11) assigned to each frame i The average value is taken as the installation deviation Δθ of the millimeter wave radar (11).

4. A method according to any one of the preceding claims, wherein: Each frame of radar point cloud data acquired by the millimeter wave radar (11) is distributed in each angle interval of the horizontal field of view angle and / or the vertical field of view angle of the millimeter wave radar (11), wherein each angle interval occupies an angle between 1° and 10°, in particular 1°.

5. A method according to any one of the preceding claims, wherein: When the valid static point cloud data is radar point cloud data acquired in the horizontal field of view of the millimeter wave radar (11), the azimuth angle installation deviation of the millimeter wave radar (11) is calculated based on the scale parameter of the valid static point cloud data by means of the Taylor series expansion of the cosine function; and / or When the valid static point cloud data is radar point cloud data acquired in the vertical field of view of the millimeter wave radar (11), the pitch angle installation deviation of the millimeter wave radar (11) is calculated based on the scale parameter of the valid static point cloud data with the aid of the Taylor series expansion of the cosine function.

6. A method according to any one of the preceding claims, wherein: Based on the scale parameter of the effective static point cloud data, the azimuth installation deviation and / or the elevation installation deviation Δθ of the millimeter wave radar (11) is calculated by using the quadratic function approximation of the Taylor series expansion of the cosine function, wherein the calculation process is described as the following equation: Wherein, θ′ represents the angle value of the angle interval in which the valid static point cloud data is located in the horizontal field of view angle and / or the vertical field of view angle of the millimeter wave radar (11).

7. A method according to any one of the preceding claims, wherein: The predetermined driving condition of the vehicle (10) includes: a speed v of the vehicle (10); h within a predetermined speed range, the predetermined speed range is, for example, 15 km / h to 120 km / h; and / or the yaw angle change rate of the vehicle (10) is within a predetermined yaw angle change rate range, the predetermined yaw angle change rate range is, for example, less than 0.03 rad / s; and / or the acceleration of the vehicle (10) is within a predetermined acceleration range, the predetermined yaw angle change rate range is, for example, less than ±5 m / s 2 .

8. A radar system (1), wherein: The radar system (1) comprises a millimeter wave radar (11) and a controller (12) for executing the method according to any one of the preceding claims.

9. A vehicle (10) comprising a radar system (1) according to claim 8. 10 . A computer program product, such as a computer-readable program carrier, comprising or storing computer program instructions, which, when executed by a processor, at least assist in implementing the steps of the method according to claim 1 .

Images