Radar calibration method

Abstract

The present application discloses a radar calibration method, which includes: obtaining absolute stationary point clouds from the current frame of radar point clouds according to the first velocity information of each point cloud in the current frame of radar point clouds in the vehicle body coordinate system and the second velocity information of the reference stationary point clouds in the vehicle body coordinate system; calculating the third velocity information of the reference stationary point clouds in the radar coordinate system when the loss function established by the radial velocity magnitude, azimuth angle, and pitch angle of the absolute stationary point clouds in the radar coordinate system takes the minimum value; determining the pitch mounting deviation angle and horizontal mounting deviation angle of the radar according to the first relative rotation matrix of the rotation from the second velocity information to the third velocity information and the second relative rotation matrix of the rotation from the radar coordinate system to the vehicle body coordinate system. This method can automatically determine the pitch mounting deviation angle and horizontal mounting deviation angle of the radar, without the need for manual construction of a calibration station to achieve radar calibration, thereby saving manpower and material resources and improving the calibration efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of automobiles, and more particularly, to a radar calibration method. Background Art

[0002] In recent years, intelligent driving technology has developed rapidly and attracted more and more attention. Environmental perception, path planning, and decision-making control are the three main modules of intelligent driving technology. Among them, the environmental perception module is the basis for the other two modules. Only with good environmental perception can better path planning be obtained and correct decisions be made. Currently, environmental perception mainly relies on three sensors: cameras, lidar, and millimeter-wave radars. Due to the characteristics of all-weather operation and less influence by the environment, millimeter-wave radars have gradually become an indispensable sensor for intelligent driving. Moreover, a 4D millimeter-wave imaging radar with high azimuth and elevation angle resolutions has been introduced, which can generate higher-quality point cloud data.

[0003] After a vehicle equipped with a millimeter-wave radar has been driving for some time, the millimeter-wave radar may become skewed. In this case, the driver needs to find after-sales to set up a calibration station to calibrate the millimeter-wave radar based on the calibration station. However, setting up and using a calibration station requires a large amount of manpower and material resources. It is suitable for batch calibration, but for a small number of calibration tasks, it will cause a waste of manpower and material resources. Summary of the Invention

[0004] The present application provides a radar calibration method, which can solve the problem of waste of manpower and material resources caused by the need to manually set up a calibration station to achieve radar calibration in the related art.

[0005] The specific technical solutions are as follows:

[0006] In a first aspect, an embodiment of the present application provides a radar calibration method, and the method includes:

[0007] Obtain absolute stationary point clouds from the current frame of radar point clouds according to the first velocity information of each point cloud in the current frame of radar point clouds in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system;

[0008] Calculate the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value, where the loss function is established based on the radial velocity magnitude, azimuth angle, and elevation angle of the absolute stationary point cloud in the radar coordinate system;

[0009] Determine the pitch installation deviation angle and the horizontal installation deviation angle of the radar according to the first relative rotation matrix of the rotation from the second velocity information to the third velocity information and the second relative rotation matrix of the rotation from the radar coordinate system to the vehicle body coordinate system.

[0010] In one embodiment, the first speed information includes the magnitude of the speed of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system, and the second speed information includes the magnitude of the speed of the reference stationary point cloud on the y-axis of the vehicle body coordinate system;

[0011] Obtaining the absolute stationary point cloud from the current frame of radar point cloud according to the first speed information of each point cloud in the current frame of radar point cloud in the vehicle body coordinate system and the second speed information of the reference stationary point cloud in the vehicle body coordinate system includes:

[0012] Calculating the ratio of the magnitude of the speed of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system to the magnitude of the speed of the reference stationary point cloud on the y-axis of the vehicle body coordinate system respectively;

[0013] Determining the point cloud in the current frame of radar point cloud with a ratio less than a preset ratio threshold as the absolute stationary point cloud, and obtaining the absolute stationary point cloud from the current frame of radar point cloud.

[0014] In one embodiment, the second speed information further includes the magnitude of the speed of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system;

[0015] Before calculating the ratio of the magnitude of the speed of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system to the magnitude of the speed of the reference stationary point cloud on the y-axis of the vehicle body coordinate system respectively, the method further includes:

[0016] Calculating the magnitude of the speed of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the vehicle speed corresponding to the current frame of radar point cloud, the angular velocity of the vehicle in the vehicle body coordinate system, and the deviation of the radar coordinate system relative to the vehicle body coordinate system;

[0017] Obtaining the magnitude of the speed of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system by taking the opposite of the magnitude of the speed of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system respectively;

[0018] For the point cloud to be calculated in the current frame of radar point cloud, calculating the magnitude of the speed of the point cloud to be calculated on the y-axis of the vehicle body coordinate system according to the azimuth angle of the point cloud to be calculated in the radar coordinate system, the pitch angle of the point cloud to be calculated in the radar coordinate system, the radial velocity of the point cloud to be calculated in the radar coordinate system, and the magnitude of the speed of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system.

[0019] In one implementation, calculating the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system based on the vehicle speed corresponding to the current frame of radar point cloud, the angular velocity of the vehicle in the vehicle body coordinate system, and the deviation of the radar coordinate system relative to the vehicle body coordinate system, includes:

[0020] Calculating the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the following formula:

[0021]

[0022]

[0023]

[0024] Wherein, and respectively represent the velocity magnitudes of the radar on the x-axis, y-axis, and z of the vehicle body coordinate system, ω x 、ω y and ω z respectively represent the angular velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, RadarX, RadarY, and RadarZ respectively represent the deviations of the radar coordinate system relative to the vehicle body coordinate system on the x-axis, y-axis, and z-axis, and VehicleSpeed represents the speed of the vehicle when generating the current frame of radar point cloud.

[0025] In one implementation, for the point cloud to be calculated in the current frame of radar point cloud, calculating the velocity magnitude of the point cloud to be calculated on the y-axis of the vehicle body coordinate system based on the azimuth angle of the point cloud to be calculated in the radar coordinate system, the pitch angle of the point cloud to be calculated in the radar coordinate system, the radial velocity of the point cloud to be calculated in the radar coordinate system, and the velocity magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, includes:

[0026] Calculating the velocity magnitude of the i-th point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system according to the following formula

[0027]

[0028] Wherein, represents the radial velocity magnitude of the i-th point cloud in the radar coordinate system, respectively represent the velocity magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, θ i represents the azimuth angle of the i-th point cloud in the radar coordinate system, represents the pitch angle of the i-th point cloud in the radar coordinate system.

[0029] In one embodiment, the third velocity information includes the magnitudes of the velocities of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system. The loss function is established based on the least squares method, and the loss function is:

[0030]

[0031] where θ1, θ2... θ N respectively represent the azimuth angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, respectively represent the elevation angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, and respectively represent the magnitudes of the velocities of the point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system. When the loss function takes the minimum value, and respectively represent the magnitudes of the velocities of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system, respectively represent the radial velocity magnitudes of each absolutely stationary point cloud in the radar coordinate system, and N is the number of absolutely stationary point clouds in the current frame of radar point cloud;

[0032] The radial velocity magnitude of the j-th absolutely stationary point cloud in the radar coordinate system is calculated according to the following formula

[0033]

[0034] where θ j represents the azimuth angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system, represents the elevation angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system.

[0035] In one embodiment, the second velocity information includes the magnitudes of the velocities of the reference stationary point cloud on the x-axis, y-axis, and z of the vehicle body coordinate system, and the third velocity information includes the magnitudes of the velocities of the reference stationary point cloud on the x-axis, y-axis, and z of the radar coordinate system;

[0036] The calculation method of the first relative rotation matrix includes:

[0037] Calculate the included angle between the first velocity vector and the second velocity vector, where the first velocity vector is generated based on the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the vehicle body coordinate system in the second velocity information, and the second velocity vector is generated based on the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system in the third velocity information;

[0038] Calculate the skew-symmetric matrix of the unit rotation axis of the first velocity vector and the second velocity vector;

[0039] Calculate the first relative rotation matrix according to the Rodriguez rotation formula, the included angle, and the skew-symmetric matrix;

[0040] In one embodiment,

[0041] The second relative rotation matrix

[0042] where, θ x is the rotation angle about the x-axis when the radar coordinate system rotates to the vehicle body coordinate system, and θ z is the rotation angle about the z-axis when the radar coordinate system rotates to the vehicle body coordinate system.

[0043] In one embodiment, obtaining the absolute stationary point cloud from the current frame of radar point cloud according to the first velocity information of each point cloud in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system in the current frame of radar point cloud includes: when the vehicle speed corresponding to the current frame of radar point cloud is greater than a preset vehicle speed threshold, obtaining the absolute stationary point cloud from the current frame of radar point cloud according to the first velocity information of each point cloud in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system;

[0044] And / or, calculating the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value, includes: when the number of absolute stationary point clouds in the current frame of radar point cloud is greater than a preset number threshold, calculating the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value.

[0045] In one embodiment, the method further includes:

[0046] Generating a first vector based on M pitch mounting offset angles corresponding to M consecutive frames of radar point clouds, and generating a second vector based on M horizontal mounting offset angles corresponding to the M consecutive frames of radar point clouds, where M is a positive integer;

[0047] Perform histogram statistics on the first vector and the second vector to obtain at least one mode of the pitch mounting deviation angle and at least one mode of the horizontal mounting deviation angle;

[0048] Determine the average value of at least one mode of the pitch mounting deviation angle as the finally required pitch mounting deviation angle, and determine the average value of at least one mode of the horizontal mounting deviation angle as the finally required horizontal mounting deviation angle.

[0049] In a second aspect, another embodiment of the present application provides a radar calibration device, and the device includes:

[0050] An acquisition unit, configured to obtain absolute stationary point clouds from the current frame of radar point cloud according to the first speed information of each point cloud in the current frame of radar point cloud in the vehicle body coordinate system and the second speed information of the reference stationary point cloud in the vehicle body coordinate system;

[0051] A calculation unit, configured to calculate the third speed information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value, where the loss function is established according to the radial speed magnitude, azimuth angle, and pitch angle of the absolute stationary point cloud in the radar coordinate system;

[0052] A determination unit, configured to determine the pitch mounting deviation angle and the horizontal mounting deviation angle of the radar according to the first relative rotation matrix of the rotation from the second speed information to the third speed information and the second relative rotation matrix of the rotation from the radar coordinate system to the vehicle body coordinate system.

[0053] In one implementation manner, the first speed information includes the speed magnitude of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system, and the second speed information includes the speed magnitude of the reference stationary point cloud on the y-axis of the vehicle body coordinate system;

[0054] The acquisition unit includes:

[0055] A calculation module, configured to calculate the ratio of the speed magnitude of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system to the speed magnitude of the reference stationary point cloud on the y-axis of the vehicle body coordinate system respectively;

[0056] A determination and acquisition module, configured to determine the point clouds in the current frame of radar point cloud with a ratio less than a preset ratio threshold as absolute stationary point clouds, and obtain absolute stationary point clouds from the current frame of radar point cloud.

[0057] In one implementation manner, the second speed information further includes the speed magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system;

[0058] The calculation module is further configured to calculate the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the vehicle speed corresponding to the current frame of radar point cloud, the angular velocity of the vehicle in the vehicle body coordinate system, and the deviation of the radar coordinate system relative to the vehicle body coordinate system, before respectively calculating the ratio of the velocity magnitude of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system to the velocity magnitude of the reference stationary point cloud on the y-axis of the vehicle body coordinate system;

[0059] The obtaining unit further includes:

[0060] The negation module is configured to obtain the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system by respectively negating the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system;

[0061] The calculation module is further configured to calculate, for the point cloud to be calculated in the current frame of radar point cloud, the velocity magnitude of the point cloud to be calculated on the y-axis of the vehicle body coordinate system according to the azimuth angle of the point cloud to be calculated in the radar coordinate system, the pitch angle of the point cloud to be calculated in the radar coordinate system, the radial velocity of the point cloud to be calculated in the radar coordinate system, and the velocity magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system.

[0062] In one embodiment, the calculation module is configured to

[0063] calculate the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the following formula:

[0064]

[0065]

[0066]

[0067] where and respectively represent the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, ω x 、ω y and ω z respectively represent the angular velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, RadarX, RadarY, and RadarZ respectively represent the deviations of the radar coordinate system relative to the vehicle body coordinate system on the x-axis, y-axis, and z-axis, and VehicleSpeed represents the speed of the vehicle when generating the current frame of radar point cloud.

[0068] In one embodiment, the calculation module is configured to

[0069] Calculate the magnitude of the velocity of the i-th point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system according to the following formula

[0070]

[0071] where represents the magnitude of the radial velocity of the i-th point cloud in the radar coordinate system, respectively represent the magnitudes of the velocities of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, and θ i represents the azimuth angle of the i-th point cloud in the radar coordinate system, represents the pitch angle of the i-th point cloud in the radar coordinate system.

[0072] In one embodiment, the third velocity information includes the magnitudes of the velocities of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system. The loss function is established based on the least squares method, and the loss function is:

[0073]

[0074] where θ1, θ2... θ N respectively represent the azimuth angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, respectively represent the pitch angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, and respectively represent the magnitudes of the velocities of the point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system. And when the loss function takes the minimum value, and respectively represent the magnitudes of the velocities of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system, respectively represent the magnitudes of the radial velocities of each absolutely stationary point cloud in the radar coordinate system, and N is the number of absolutely stationary point clouds in the current frame of radar point cloud;

[0075] A calculation unit for calculating the magnitude of the radial velocity of the j-th absolutely stationary point cloud in the radar coordinate system according to the following formula

[0076]

[0077] where θ j represents the azimuth angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system, represents the pitch angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system.

[0078] In one embodiment, the second speed information includes the magnitudes of the speeds of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, and the third speed information includes the magnitudes of the speeds of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system;

[0079] The determining unit is further configured to calculate the included angle between the first speed vector and the second speed vector, where the first speed vector is generated according to the magnitudes of the speeds of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the vehicle body coordinate system in the second speed information, and the second speed vector is generated according to the magnitudes of the speeds of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system in the third speed information; calculate the skew-symmetric matrix of the unit rotation axis of the first speed vector and the second speed vector; and calculate the first relative rotation matrix according to the Rodriguez rotation formula, the included angle, and the skew-symmetric matrix.

[0080] In one embodiment,

[0081] The second relative rotation matrix

[0082] where θ x is the rotation angle about the x-axis when the radar coordinate system rotates to the vehicle body coordinate system, and θ z is the rotation angle about the z-axis when the radar coordinate system rotates to the vehicle body coordinate system.

[0083] In one embodiment, the obtaining unit is configured to, when the vehicle speed corresponding to the current frame of radar point cloud is greater than a preset vehicle speed threshold, obtain the absolute stationary point cloud from the current frame of radar point cloud according to the first speed information of each point cloud in the vehicle body coordinate system of the current frame of radar point cloud and the second speed information of the reference stationary point cloud in the vehicle body coordinate system;

[0084] and / or,

[0085] The calculating unit is configured to, when the number of absolute stationary point clouds in the current frame of radar point cloud is greater than a preset number threshold, calculate the third speed information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value.

[0086] In one embodiment, the apparatus further includes:

[0087] The generating unit is configured to generate a first vector based on M pitch mounting deflection angles corresponding to M consecutive frames of radar point clouds, and generate a second vector based on M horizontal mounting deflection angles corresponding to the M consecutive frames of radar point clouds, where M is a positive integer;

[0088] A statistical unit for performing histogram statistics on the first vector and the second vector to obtain at least one mode of the pitch mounting offset angle and at least one mode of the horizontal mounting offset angle;

[0089] A determination unit is further configured to determine the average value of at least one mode of the pitch mounting offset angle as the finally required pitch mounting offset angle, and determine the average value of at least one mode of the horizontal mounting offset angle as the finally required horizontal mounting offset angle.

[0090] In a third aspect, another embodiment of the present application provides a storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor implements the method according to any one of the embodiments in the first aspect.

[0091] In a fourth aspect, another embodiment of the present application provides an electronic device, the electronic device includes:

[0092] One or more processors;

[0093] A storage device for storing one or more programs,

[0094] wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of the embodiments in the first aspect.

[0095] As can be seen from the above, the radar calibration method provided by the embodiments of the present application can first obtain the absolute stationary point cloud from the current frame of radar point cloud according to the first velocity information of each point cloud in the current frame of radar point cloud in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system, and then calculate the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function established by the radial velocity magnitude, azimuth angle and pitch angle of the absolute stationary point cloud in the radar coordinate system takes the minimum value. Finally, according to the first relative rotation matrix of the second velocity information rotating to the third velocity information and the second relative rotation matrix of the radar coordinate system rotating to the vehicle body coordinate system, the pitch mounting offset angle and the horizontal mounting offset angle of the radar are determined. It can be seen that the embodiments of the present application can calculate the pitch mounting offset angle and the horizontal mounting offset angle of the radar according to relevant information such as radar data and inertial navigation information during the vehicle driving process, without manually building a calibration station to implement radar calibration, thereby not only saving manpower and material resources, but also improving the calibration efficiency.

[0096] The technical effects that can be achieved by the embodiments of the present application also include but are not limited to the following points:

[0097] 1. Since different vehicle speeds have different effects on radar calibration, in order to improve the accuracy of radar calibration, the pitch installation deviation angle and the horizontal installation deviation angle corresponding to the current frame of radar point cloud can be calculated only when the vehicle speed corresponding to the current frame of radar point cloud is greater than the preset vehicle speed threshold.

[0098] 2. In order to improve the accuracy of the third speed information, after obtaining the absolutely stationary point cloud, it can be determined whether the number of absolutely stationary point clouds is greater than the preset number threshold. Only when the number of absolutely stationary point clouds is greater than the preset number threshold, the pitch installation deviation angle and the horizontal installation deviation angle corresponding to the current frame of radar point cloud are calculated.

[0099] 3. In order to eliminate the random error of the system, the pitch installation deviation angle and the horizontal installation deviation angle of the continuous M frames of radar point cloud can be statistically analyzed to obtain the final required pitch installation deviation angle and horizontal installation deviation angle, thereby improving the accuracy of radar calibration.

[0100] Of course, it is not necessary for any product or method implementing the present application to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0102] Figure 1 It is a schematic flowchart of a radar calibration method provided by an embodiment of the present application;

[0103] Figure 2 It is a schematic flowchart of another radar calibration method provided by an embodiment of the present application;

[0104] Figure 3 It is a schematic diagram of a coordinate system provided by an embodiment of the present application;

[0105] Figure 4 It is a schematic flowchart of yet another radar calibration method provided by an embodiment of the present application;

[0106] Figure 5 It is a block diagram of the composition of a radar calibration device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0107] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0108] It should be noted that the terms "including" and "having" and any variations thereof in the embodiments of the present application and the accompanying drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include steps or units not listed, or may optionally further include other steps or units inherent to these processes, methods, products, or devices.

[0109] Figure 1 The following is a schematic flowchart of a radar calibration method provided by an embodiment of the present application. This method can be applied to the vehicle side or the server side. When applied to the server side, the vehicle can transmit relevant data to the server for processing. This method mainly includes:

[0110] S110: Obtain absolute stationary point clouds from the current frame of radar point cloud according to the first velocity information of each point cloud in the current frame of radar point cloud in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system.

[0111] The first velocity information includes the velocity magnitude of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system, and the second velocity information includes the velocity magnitude of the reference stationary point cloud on the y-axis of the vehicle body coordinate system.

[0112] As Figure 2 shown, the specific implementation manner of this step may include steps S111 - S112:

[0113] (S111) Calculate the ratio of the velocity magnitude of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system to the velocity magnitude of the reference stationary point cloud on the y-axis of the vehicle body coordinate system respectively.

[0114] The second velocity information further includes the velocity magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system.

[0115] The implementation process of calculating the velocity magnitude of each point cloud on the y-axis of the vehicle body coordinate system and the velocity magnitude of the reference stationary point cloud on the y-axis of the vehicle body coordinate system includes steps A01 - A03:

[0116] (A01) Calculate the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system based on the vehicle speed, the angular velocity of the vehicle in the vehicle body coordinate system, and the deviation of the radar coordinate system relative to the vehicle body coordinate system corresponding to the current frame of radar point cloud.

[0117] Calculate the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the following formula:

[0118]

[0119]

[0120]

[0121] Wherein, and respectively represent the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, ω x , ω y and ω z respectively represent the angular velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, RadarX, RadarY, and RadarZ respectively represent the deviations of the radar coordinate system relative to the vehicle body coordinate system on the x-axis, y-axis, and z-axis, and VehicleSpeed represents the speed of the vehicle when generating the current frame of radar point cloud. Figure 3 is a schematic diagram of the radar coordinate system and the vehicle body coordinate system, showing the x-axis and y-axis information (the z-axis is not shown) of the radar coordinate system and the vehicle body coordinate system, as well as the deviations RadarX and RadarY of the two on the x-axis and y-axis. In addition, the vehicle speed and the angular velocity of the vehicle in the vehicle body coordinate system can be measured by an IMU (Inertial Measurement Unit).

[0122] (A02) After taking the opposite of the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system respectively, obtain the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system.

[0123] The relative motion vector of the reference stationary point cloud in the vehicle body coordinate system is opposite to the motion vector of the radar. Therefore, the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system can be obtained by taking the opposite of the velocity magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system respectively. It is expressed by the formula as follows:

[0124]

[0125]

[0126]

[0127] Among them, respectively represent the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system.

[0128] (A03) For the point cloud to be calculated in the current frame of radar point cloud, according to the azimuth angle of the point cloud to be calculated in the radar coordinate system, the pitch angle of the point cloud to be calculated in the radar coordinate system, the radial velocity of the point cloud to be calculated in the radar coordinate system, and the velocity magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, calculate the velocity magnitude of the point cloud to be calculated on the y-axis of the vehicle body coordinate system.

[0129] Due to the small installation angle, the influence on is small, so the velocity magnitude of the i-th point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system can be calculated according to the following formula

[0130]

[0131] Among them, represents the radial velocity magnitude of the i-th point cloud in the radar coordinate system, respectively represent the velocity magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, and θ i represents the azimuth angle of the i-th point cloud in the radar coordinate system, represents the pitch angle of the i-th point cloud in the radar coordinate system.

[0132]

[0133] Among them, and respectively represent the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system, and θ i represents the azimuth angle of the i-th point cloud in the current frame of radar point cloud in the radar coordinate system, represents the pitch angle of the i-th point cloud in the current frame of radar point cloud in the radar coordinate system.

[0134] (S112) Determine the absolutely stationary point cloud from the point clouds in the current frame of radar point cloud whose ratio is less than the preset ratio threshold, and obtain the absolutely stationary point cloud from the current frame of radar point cloud.

[0135] Among them, the preset ratio threshold is determined according to the radar characteristics and actual experience, as long as the final radar calibration accuracy meets the requirements, for example, it can be 1000.

[0136] Since the angular deviation between the radar coordinate system and the vehicle body coordinate system is relatively small, the speed magnitude of each point cloud on the x-axis of the vehicle body coordinate system can be considered equal to that of the reference stationary point cloud on the x-axis of the vehicle body coordinate system, and the speed magnitude of each point cloud on the y-axis of the vehicle body coordinate system can also be considered equal to that of the reference stationary point cloud on the y-axis of the vehicle body coordinate system. Therefore, by comparing the speed magnitude of each point cloud on the y-axis of the vehicle body coordinate system with that of the reference stationary point cloud on the y-axis of the vehicle body coordinate system, if the difference is very small, the point cloud to be compared is considered an absolutely stationary point cloud.

[0137] It should be added that when obtaining the absolutely stationary point cloud from the current frame of radar point cloud in step S110, all absolutely stationary point clouds can be obtained for subsequent calculations, or only some absolutely stationary point clouds can be obtained for subsequent calculations, as long as the number of absolutely stationary point clouds participating in the calculations can meet the requirements of subsequent calibration accuracy.

[0138] In one implementation, since different vehicle speeds have different effects on radar calibration, to improve the accuracy of radar calibration, when the vehicle speed corresponding to the current frame of radar point cloud is greater than the preset vehicle speed threshold, the absolutely stationary point cloud can be obtained from the current frame of radar point cloud according to the first speed information of each point cloud in the vehicle body coordinate system and the second speed information of the reference stationary point cloud in the vehicle body coordinate system, and when the vehicle speed is less than or equal to the preset vehicle speed threshold, the operation of obtaining the absolutely stationary point cloud from the current frame of radar point cloud according to the first speed information of each point cloud in the vehicle body coordinate system and the second speed information of the reference stationary point cloud in the vehicle body coordinate system is not performed. Among them, the preset vehicle speed threshold is determined according to actual test experience and can be, for example, 5m / s.

[0139] S120: Calculate the third speed information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value.

[0140] The loss function is established based on the radial speed magnitude, azimuth angle, and pitch angle of the absolutely stationary point cloud in the radar coordinate system. The third speed information includes the speed magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system. The loss function is established based on the least squares method, and the loss function is:

[0141]

[0142] where θ1, θ2…θ N respectively represent the azimuth angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, respectively represent the pitch angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, and respectively represent the velocity magnitudes of the point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system, and when the loss function takes the minimum value, and respectively represent the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system, respectively represent the radial velocity magnitudes of each absolutely stationary point cloud in the radar coordinate system, and N is the number of absolutely stationary point clouds in the current frame of radar point cloud.

[0143] Calculate the radial velocity magnitude of the j-th absolutely stationary point cloud in the radar coordinate system according to the following formula

[0144] θ j represents the azimuth angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system, represents the pitch angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system.

[0145] In one implementation, since the more the number of absolutely stationary point clouds, the higher the accuracy of the third velocity information calculated using the loss function, in order to improve the accuracy of the third velocity information, after obtaining the absolutely stationary point clouds, it can be determined whether the number of absolutely stationary point clouds is greater than a preset number threshold. When the number of absolutely stationary point clouds in the current frame of radar point cloud is greater than the preset number threshold, step S120 is executed; otherwise, step S120 is not executed. Among them, the preset number threshold is determined according to the radar calibration accuracy requirements.

[0146] S130: Determine the pitch mounting offset angle and horizontal mounting offset angle of the radar according to the first relative rotation matrix for the rotation from the second velocity information to the third velocity information and the second relative rotation matrix for the rotation from the radar coordinate system to the vehicle body coordinate system.

[0147] The second velocity information includes the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z of the vehicle body coordinate system, and the third velocity information includes the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z of the radar coordinate system.

[0148] In one implementation, the calculation method of the first relative rotation matrix includes:

[0149] (B1) Calculate the angle between the first velocity vector and the second velocity vector, where the first velocity vector is generated according to the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z of the vehicle body coordinate system in the second velocity information, and the second velocity vector is generated according to the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z of the radar coordinate system in the third velocity information.

[0150] Assume the first velocity vector The second velocity vector The method for calculating the included angle between the first velocity vector and the second velocity vector is as follows:

[0151]

[0152] where α is the included angle.

[0153] (B2) Calculate the skew-symmetric matrix of the unit rotation axis of the first velocity vector and the second velocity vector.

[0154] According to the following formula for the unit rotation axis of the first velocity vector and the second velocity vector:

[0155]

[0156] where ω is the unit rotation axis.

[0157] Let ω = [ω1 ω2 ω3] T , and its skew-symmetric matrix is

[0158]

[0159] (B3) Calculate the first relative rotation matrix according to the Rodriguez rotation formula, the included angle, and the skew-symmetric matrix.

[0160] According to the Rodriguez rotation formula, the rotation matrix can be deduced as:

[0161]

[0162] where I is the 3×3 identity matrix.

[0163] In one implementation, the angular deviation of the radar coordinate system relative to the vehicle body coordinate system only exists in the azimuth direction (rotation around the z-axis) and the pitch direction (rotation around the x-axis). Therefore, the calculation method of its general rotation matrix, i.e., the second relative rotation matrix, includes:

[0164] The second relative rotation matrix

[0165] where θ x is the rotation angle around the x-axis when the radar coordinate system rotates to the vehicle body coordinate system, and θ z is the rotation angle around the z-axis when the radar coordinate system rotates to the vehicle body coordinate system.

[0166] After obtaining the first relative rotation matrix and the second relative rotation matrix, the pitch mounting deviation angle and the horizontal mounting deviation angle of the radar can be calculated according to the formula ​

[0167] The radar calibration method provided by the embodiments of the present application can first obtain the absolutely stationary point cloud from the current frame of radar point cloud according to the first velocity information of each point cloud in the current frame of radar point cloud in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system, and then calculate the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function established by the radial velocity magnitude, azimuth angle, and pitch angle of the absolutely stationary point cloud in the radar coordinate system takes the minimum value. Finally, according to the first relative rotation matrix for rotating from the second velocity information to the third velocity information and the second relative rotation matrix for rotating from the radar coordinate system to the vehicle body coordinate system, the pitch mounting deviation angle and the horizontal mounting deviation angle of the radar are determined. It can be seen that the embodiments of the present application can calculate the pitch mounting deviation angle and the horizontal mounting deviation angle of the radar according to relevant information such as radar data and inertial navigation information during the vehicle driving process, without manually building a calibration station to implement radar calibration, thus not only saving manpower and material resources, but also improving the calibration efficiency.

[0168] In one implementation, since there are random errors in the mounting deviation angle solved by a single frame, in order to eliminate the random errors of the system, as Figure 4 shown, the embodiments of the present application provide the following method:

[0169] S210: Obtain the absolutely stationary point cloud from the current frame of radar point cloud according to the first velocity information of each point cloud in the current frame of radar point cloud in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system.

[0170] S220: Calculate the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value.

[0171] S230: Determine the pitch mounting deviation angle and the horizontal mounting deviation angle of the radar according to the first relative rotation matrix for rotating from the second velocity information to the third velocity information and the second relative rotation matrix for rotating from the radar coordinate system to the vehicle body coordinate system.

[0172] S240: Generate a first vector based on the M pitch mounting deviation angles corresponding to the continuous M frames of radar point clouds, and generate a second vector based on the M horizontal mounting deviation angles corresponding to the continuous M frames of radar point clouds.

[0173] Wherein, the M is a positive integer.

[0174] S250: Perform histogram statistics on the first vector and the second vector to obtain at least one mode of the pitch mounting deviation angle and at least one mode of the horizontal mounting deviation angle.

[0175] Wherein, the first vector θ x =[θ x1 θ x2… θ xm , the second vector θ z = [θ z1 θ z2 … θ zm .

[0176] S260: Determine the average value of at least one mode of the pitch installation deviation angle as the finally required pitch installation deviation angle, and determine the average value of at least one mode of the horizontal installation deviation angle as the finally required horizontal installation deviation angle.

[0177] The mode is the value that appears most frequently in a set of data. Sometimes there may be multiple values that appear most frequently in a set of data. Therefore, the finally required pitch installation deviation angle and horizontal installation deviation angle can be obtained by calculating the average value of at least one mode of the pitch installation deviation angle and the average value of at least one mode of the horizontal installation deviation angle.

[0178] Based on the above embodiments, an embodiment of the present application provides a radar calibration device, as Figure 5 shown, the device includes:

[0179] An acquisition unit 30, configured to obtain absolute stationary point clouds from the current frame of radar point clouds according to the first velocity information of each point cloud in the current frame of radar point clouds in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system;

[0180] A calculation unit 32, configured to calculate the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value, where the loss function is established according to the radial velocity magnitude, azimuth angle, and pitch angle of the absolute stationary point cloud in the radar coordinate system;

[0181] A determination unit 34, configured to determine the pitch installation deviation angle and the horizontal installation deviation angle of the radar according to the first relative rotation matrix of the rotation of the second velocity information to the third velocity information and the second relative rotation matrix of the rotation of the radar coordinate system to the vehicle body coordinate system.

[0182] In an implementation manner, the first velocity information includes the velocity magnitude of each point cloud in the current frame of radar point clouds on the y-axis of the vehicle body coordinate system, and the second velocity information includes the velocity magnitude of the reference stationary point cloud on the y-axis of the vehicle body coordinate system;

[0183] The acquisition unit 30 includes:

[0184] A calculation module, configured to calculate the ratio of the velocity magnitude of each point cloud in the current frame of radar point clouds on the y-axis of the vehicle body coordinate system to the velocity magnitude of the reference stationary point cloud on the y-axis of the vehicle body coordinate system respectively;

[0185] A determination and acquisition module, configured to determine the point cloud with a ratio less than a preset ratio threshold in the current frame of radar point cloud as the absolute stationary point cloud, and acquire the absolute stationary point cloud from the current frame of radar point cloud.

[0186] In one implementation, the second velocity information further includes the magnitudes of the velocities of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system;

[0187] The calculation module is further configured to, before calculating the ratio of the magnitude of the velocity of each point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system to the magnitude of the velocity of the reference stationary point cloud on the y-axis of the vehicle body coordinate system, calculate the magnitudes of the velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the vehicle speed corresponding to the current frame of radar point cloud, the angular velocity of the vehicle in the vehicle body coordinate system, and the deviation of the radar coordinate system relative to the vehicle body coordinate system;

[0188] The acquisition unit 30 further includes:

[0189] An inversion module, configured to obtain the magnitudes of the velocities of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system after respectively inverting the magnitudes of the velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system;

[0190] The calculation module is further configured to, for the point cloud to be calculated in the current frame of radar point cloud, calculate the magnitude of the velocity of the point cloud to be calculated on the y-axis of the vehicle body coordinate system according to the azimuth angle of the point cloud to be calculated in the radar coordinate system, the pitch angle of the point cloud to be calculated in the radar coordinate system, the radial velocity of the point cloud to be calculated in the radar coordinate system, and the magnitudes of the velocities of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system.

[0191] In one implementation, the calculation module is configured to

[0192] Calculate the magnitudes of the velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the following formula:

[0193]

[0194]

[0195]

[0196] Wherein, and respectively represent the magnitudes of the velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, ω x 、ω y and ω zrespectively represent the angular velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, RadarX, RadarY, and RadarZ respectively represent the deviations of the radar coordinate system relative to the vehicle body coordinate system on the x-axis, y-axis, and z-axis, and VehicleSpeed represents the speed of the vehicle when generating the current frame of radar point cloud.

[0197] In one implementation, a calculation module is used for

[0198] Calculate the speed magnitude of the i-th point cloud in the current frame of radar point cloud on the y-axis of the vehicle body coordinate system according to the following formula

[0199]

[0200] Where represents the radial speed magnitude of the i-th point cloud in the radar coordinate system, respectively represent the speed magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, and θ i represents the azimuth angle of the i-th point cloud in the radar coordinate system, represents the pitch angle of the i-th point cloud in the radar coordinate system.

[0201] In one implementation, the third speed information includes the speed magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system, and the loss function is established based on the least squares method. The loss function is:

[0202]

[0203] Where θ1, θ2…θ N respectively represent the azimuth angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, respectively represent the pitch angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, and respectively represent the speed magnitudes of the point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system, and when the loss function takes the minimum value, and respectively represent the speed magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system, respectively represent the radial speed magnitudes of each absolutely stationary point cloud in the radar coordinate system, and N is the number of absolutely stationary point clouds in the current frame of radar point cloud;

[0204] The calculation unit 32 is used to calculate the radial speed magnitude of the j-th absolutely stationary point cloud in the radar coordinate system according to the following formula

[0205]

[0206] Among them, θ j represents the azimuth angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system, and represents the pitch angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system.

[0207] In one implementation, the second velocity information includes the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the vehicle body coordinate system, and the third velocity information includes the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system;

[0208] The determination unit 34 is further configured to calculate the included angle between the first velocity vector and the second velocity vector, where the first velocity vector is generated according to the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the vehicle body coordinate system in the second velocity information, and the second velocity vector is generated according to the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system in the third velocity information; calculate the skew-symmetric matrix of the unit rotation axis of the first velocity vector and the second velocity vector; and calculate the first relative rotation matrix according to the Rodriguez rotation formula, the included angle, and the skew-symmetric matrix.

[0209] In one implementation,

[0210] The second relative rotation matrix

[0211] Among them, θ x is the rotation angle around the x-axis when the radar coordinate system rotates to the vehicle body coordinate system, and θ z is the rotation angle around the z-axis when the radar coordinate system rotates to the vehicle body coordinate system.

[0212] In one implementation, the acquisition unit 30 is configured to, when the vehicle speed corresponding to the current frame of radar point cloud is greater than a preset vehicle speed threshold, obtain the absolutely stationary point cloud from the current frame of radar point cloud according to the first velocity information of each point cloud in the vehicle body coordinate system in the current frame of radar point cloud and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system;

[0213] and / or,

[0214] The calculation unit 32 is configured to, when the number of the absolutely stationary point clouds is greater than a preset number threshold, calculate the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value.

[0215] In one embodiment, the apparatus further comprises:

[0216] a generating unit, configured to generate a first vector based on M pitch mounting declination angles corresponding to M consecutive frames of radar point clouds, and generate a second vector based on M horizontal mounting declination angles corresponding to the M consecutive frames of radar point clouds, where M is a positive integer;

[0217] a statistical unit, configured to perform histogram statistics on the first vector and the second vector to obtain at least one mode of the pitch mounting declination angle and at least one mode of the horizontal mounting declination angle;

[0218] a determining unit 34, further configured to determine an average value of at least one mode of the pitch mounting declination angle as the finally required pitch mounting declination angle, and determine an average value of at least one mode of the horizontal mounting declination angle as the finally required horizontal mounting declination angle.

[0219] Based on the above method embodiments, another embodiment of the present application provides a storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor implements the method described in any one of the above method embodiments.

[0220] Based on the above method embodiments, another embodiment of the present application provides an electronic device, the electronic device comprising: one or more processors;

[0221] a storage device, configured to store one or more programs,

[0222] wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any one of the above method embodiments.

[0223] The above system and apparatus embodiments correspond to the method embodiments, and have the same technical effects as the method embodiments. For specific descriptions, please refer to the method embodiments. The apparatus embodiments are obtained based on the method embodiments. For specific descriptions, please refer to the method embodiment part, and details are not described herein again. Those of ordinary skill in the art can understand that: the drawings are only schematic diagrams of one embodiment, and the modules or processes in the drawings are not necessarily essential for implementing the present application.

[0224] Those of ordinary skill in the art can understand that: the modules in the apparatus in the embodiments can be distributed in the apparatus in the embodiments according to the descriptions in the embodiments, or can be correspondingly changed to be located in one or more apparatuses different from the present embodiments. The modules in the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0225] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A radar calibration method, characterized in that, The method includes: Obtaining absolute stationary point clouds from the current frame of lidar point cloud according to the first velocity information of each point cloud in the current frame of lidar point cloud in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system. When obtaining the absolute stationary point clouds from the current frame of lidar point cloud, the first velocity information includes the magnitude of the velocity of each point cloud in the current frame of lidar point cloud on the y-axis of the vehicle body coordinate system, and the second velocity information includes the magnitude of the velocity of the reference stationary point cloud on the y-axis of the vehicle body coordinate system; Calculating the third velocity information of the reference stationary point cloud in the lidar coordinate system when the loss function takes the minimum value, where the loss function is established based on the radial velocity magnitude, azimuth angle, and pitch angle of the absolute stationary point cloud in the lidar coordinate system; Determining the pitch mounting deviation angle and horizontal mounting deviation angle of the lidar according to the first relative rotation matrix of the rotation from the second velocity information to the third velocity information and the second relative rotation matrix of the rotation from the lidar coordinate system to the vehicle body coordinate system. When determining the first relative rotation matrix, the second velocity information includes the magnitude of the velocity of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the vehicle body coordinate system.

2. The method according to claim 1, wherein Obtaining absolute stationary point clouds from the current frame of lidar point cloud according to the first velocity information of each point cloud in the current frame of lidar point cloud in the vehicle body coordinate system and the second velocity information of the reference stationary point cloud in the vehicle body coordinate system, includes: Calculating the ratio of the magnitude of the velocity of each point cloud in the current frame of lidar point cloud on the y-axis of the vehicle body coordinate system to the magnitude of the velocity of the reference stationary point cloud on the y-axis of the vehicle body coordinate system respectively; Determining the point clouds in the current frame of lidar point cloud with a ratio less than a preset ratio threshold as absolute stationary point clouds, and obtaining the absolute stationary point clouds from the current frame of lidar point cloud.

3. The method according to claim 2, wherein In the case where the second velocity information includes the magnitude of the velocity of the reference stationary point cloud on the y-axis of the vehicle body coordinate system, directly execute calculating the ratio of the magnitude of the velocity of each point cloud in the current frame of lidar point cloud on the y-axis of the vehicle body coordinate system to the magnitude of the velocity of the reference stationary point cloud on the y-axis of the vehicle body coordinate system respectively; In the case where the second velocity information further includes the magnitude of the velocity of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, before calculating the ratio of the magnitude of the velocity of each point cloud in the current frame of lidar point cloud on the y-axis of the vehicle body coordinate system to the magnitude of the velocity of the reference stationary point cloud on the y-axis of the vehicle body coordinate system respectively, the method further includes: Calculating the magnitude of the velocity of the lidar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the vehicle speed corresponding to the current frame of lidar point cloud, the angular velocity of the vehicle in the vehicle body coordinate system, and the deviation of the lidar coordinate system relative to the vehicle body coordinate system; Obtaining the magnitude of the velocity of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the lidar coordinate system by taking the opposite of the magnitude of the velocity of the lidar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system respectively; For the point cloud to be calculated in the current frame radar point cloud, calculate the speed magnitude of the point cloud to be calculated on the y-axis of the vehicle body coordinate system according to the azimuth angle of the point cloud to be calculated in the radar coordinate system, the pitch angle of the point cloud to be calculated in the radar coordinate system, the radial velocity of the point cloud to be calculated in the radar coordinate system, and the speed magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system.

4. The method according to claim 3, characterized in that, Calculate the speed magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the vehicle speed corresponding to the current frame radar point cloud, the angular velocity of the vehicle in the vehicle body coordinate system, and the deviation of the radar coordinate system relative to the vehicle body coordinate system, including: Calculate the speed magnitudes of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system according to the following formula: Among them, and respectively represent the magnitudes of the speeds of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system. ω x , ω y and ω z respectively represent the angular velocities of the radar on the x-axis, y-axis, and z-axis of the vehicle body coordinate system. RadarX, RadarY, and RadarZ respectively represent the deviations of the radar coordinate system relative to the vehicle body coordinate system on the x-axis, y-axis, and z-axis. VehicleSpeed represents the speed of the vehicle when generating the current frame of radar point cloud.

5. The method according to claim 3, characterized in that, For the point cloud to be calculated in the current frame radar point cloud, calculate the speed magnitude of the point cloud to be calculated on the y-axis of the vehicle body coordinate system according to the azimuth angle of the point cloud to be calculated in the radar coordinate system, the pitch angle of the point cloud to be calculated in the radar coordinate system, the radial velocity of the point cloud to be calculated in the radar coordinate system, and the speed magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, including: Calculate the magnitude of the velocity of the \(i\)-th point cloud in the current frame of lidar point cloud on the \(y\)-axis of the vehicle body coordinate system according to the following formula Among them, represents the magnitude of the radial velocity of the i-th point cloud in the radar coordinate system, respectively represent the velocity magnitudes of the reference stationary point cloud on the x-axis and z-axis of the vehicle body coordinate system, and θ i represents the azimuth angle of the i-th point cloud in the radar coordinate system, represents the pitch angle of the i-th point cloud in the radar coordinate system.

6. The method according to claim 1, characterized in that, The third speed information includes the speed magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z-axis of the radar coordinate system. The loss function is established based on the least squares method, and the loss function is: Among them, θ1, θ2…θ N respectively represent the azimuth angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, respectively represent the pitch angles of different point clouds in the current frame of radar point cloud in the radar coordinate system, and respectively represent the velocity magnitudes of the point cloud on the x-axis, y-axis and z-axis of the radar coordinate system. And when the loss function takes the minimum value, and respectively represent the velocity magnitudes of the reference stationary point cloud on the x-axis, y-axis and z-axis of the radar coordinate system, respectively represent the radial velocity magnitudes of each absolutely stationary point cloud in the radar coordinate system, and N is the number of absolutely stationary point clouds in the current frame of radar point cloud; Calculate the radial velocity magnitude of the j-th absolutely stationary point cloud in the radar coordinate system according to the following formula Among them, θ j represents the azimuth angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system, and represents the pitch angle of the j-th absolutely stationary point cloud in the current frame of radar point cloud in the radar coordinate system.

7. The method according to claim 1, characterized in that, The third speed information includes the speed magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z of the radar coordinate system; The calculation method of the first relative rotation matrix includes: Calculate the included angle between the first speed vector and the second speed vector, where the first speed vector is generated according to the speed magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z of the vehicle body coordinate system in the second speed information, and the second speed vector is generated according to the speed magnitudes of the reference stationary point cloud on the x-axis, y-axis, and z of the radar coordinate system in the third speed information; Calculate the skew-symmetric matrix of the unit rotation axis of the first speed vector and the second speed vector; Calculate the first relative rotation matrix according to the Rodriguez rotation formula, the included angle, and the skew-symmetric matrix.

8. The method according to claim 1, wherein the second relative rotation matrix where, θ x is the rotation angle about the x-axis when the radar coordinate system rotates to the vehicle body coordinate system, and θ z is the rotation angle about the z-axis when the radar coordinate system rotates to the vehicle body coordinate system.

9. The method according to claim 1, wherein Obtain the absolutely stationary point cloud from the current frame radar point cloud according to the first speed information of each point cloud in the current frame radar point cloud in the vehicle body coordinate system and the second speed information of the reference stationary point cloud in the vehicle body coordinate system, including: when the vehicle speed corresponding to the current frame radar point cloud is greater than the preset vehicle speed threshold, obtain the absolutely stationary point cloud from the current frame radar point cloud according to the first speed information of each point cloud in the current frame radar point cloud in the vehicle body coordinate system and the second speed information of the reference stationary point cloud in the vehicle body coordinate system; And / or, calculating the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value, including: when the number of absolute stationary point clouds in the current frame radar point cloud is greater than a preset number threshold, calculating the third velocity information of the reference stationary point cloud in the radar coordinate system when the loss function takes the minimum value.

10. The method according to any one of claims 1-9, characterized in that, The method further includes: Generating a first vector based on M pitch mounting declination angles corresponding to M consecutive frames of radar point clouds, and generating a second vector based on M horizontal mounting declination angles corresponding to the M consecutive frames of radar point clouds, where M is a positive integer; Performing histogram statistics on the first vector and the second vector to obtain at least one mode of the pitch mounting declination angle and at least one mode of the horizontal mounting declination angle; Determining the average value of at least one mode of the pitch mounting declination angle as the finally required pitch mounting declination angle, and determining the average value of at least one mode of the horizontal mounting declination angle as the finally required horizontal mounting declination angle.

Images