Self-calibration processing method, system and platform for pitch angle of 4D imaging millimeter wave radar based on ground and high-altitude target

Abstract

The invention discloses a 4D imaging millimeter wave radar pitch angle self-calibration processing method, system and platform based on ground and high-altitude targets. The simple and reliable vehicle-mounted 4D imaging millimeter wave radar pitch angle self-calibration method is realized by using common flat road surfaces and high-altitude markers in vehicle driving as references, the mounting angle deviation caused by vibration and jolt can be automatically compensated in the vehicle life cycle, the industrial problem of inaccurate radar height measurement is effectively solved, and the vehicle-mounted 4D imaging millimeter wave radar pitch angle self-calibration method is suitable for popularization and application. And the sensing precision and decision reliability of the advanced driving assistance system are obviously improved.

Description

Technical Field

[0001] This invention belongs to the field of automotive electronic millimeter-wave radar technology, specifically relating to a method, system, and platform for self-calibration of elevation angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets. Background Technology

[0002] 77GHz automotive millimeter-wave radar is one of the core sensors in Advanced Driver Assistance Systems (ADAS) and autonomous driving systems. The radar's angular detection accuracy, especially its pitch angle (altitude measurement) accuracy, directly determines the performance of ADAS functions such as lane keeping, forward cross-traffic alert (FCTB), automatic emergency braking (AEB), target classification and recognition, free space perception, and grid occupancy. 4D imaging millimeter-wave radar, which emerged after 2021, adds high-precision pitch angle measurement capabilities to traditional radar ranging, speed measurement, and azimuth measurement. Its point cloud density, resolution, and measurement accuracy far exceed those of traditional radar, making it a leading industry trend.

[0003] The principle of radar altitude measurement is based on the formula H = r × sinφ, where r is the target distance and φ is the elevation angle measured by the radar. However, the installation position of the radar on a vehicle can change slightly due to factors such as long-term use and road surface vibrations, causing a deviation in the installation angle of the radar's baseline in the elevation direction from the horizontal plane. This deviation can directly lead to errors in altitude calculation, such as misjudging a manhole cover on the ground as an insurmountable obstacle, or misjudging a road sign in the air as an obstacle in front of the vehicle, seriously affecting the safety of the system's decision-making.

[0004] Currently, radars undergo end-of-line (EOL) calibration before leaving the factory to compensate for initial installation deviations. However, during a vehicle's lifespan, changes in the mounting angle due to vibrations, collisions, and other factors are unavoidable, and EOL calibration cannot solve this problem.

[0005] Therefore, in order to address the above-mentioned technical problems and deficiencies, there is an urgent need to design and develop a method, system, and platform for self-calibration of elevation angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets. Summary of the Invention

[0006] To overcome the shortcomings and difficulties of the existing technology, the purpose of this invention is to propose a method, system and platform for self-calibration of elevation angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets, so as to continuously ensure the angle measurement accuracy of the radar.

[0007] The first objective of this invention is to provide a method for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets; the second objective of this invention is to provide a system for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets; and the third objective of this invention is to provide a platform for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets.

[0008] The first objective of this invention is achieved as follows: the method comprises the following steps:

[0009] Generate and acquire first data corresponding to the real-time operating status of the vehicle and radar detection, and analyze and process the first data; wherein, the first data includes ground point cloud data and high-altitude target point cloud data;

[0010] A first change relationship corresponding to the first data is constructed, and based on the first change relationship, second data corresponding to the radar elevation angle is determined and generated; wherein, the second data is installation deviation direction data; and the first change relationship is altitude change pattern data.

[0011] Based on the second data, the radar elevation angle is iteratively compensated until the first change relationship meets the preset calibration completion conditions.

[0012] Furthermore, the step of generating and acquiring first data corresponding to the real-time vehicle status and radar detection, and analyzing and processing the first data, further includes:

[0013] Generate and acquire third data corresponding to the real-time operating status of the vehicle; wherein, the third data is data on the road surface undulation status monitored during vehicle operation;

[0014] Based on the third data, corresponding first control data is generated, and the pitch angle self-calibration program is started based on the first control data; wherein, the first control data is the control data when the preset calibration trigger condition is met.

[0015] Furthermore, the step of generating and acquiring first data corresponding to the real-time vehicle status and radar detection, and analyzing and processing the first data, further includes:

[0016] Establish calibration trigger conditions corresponding to the real-time operating status of the vehicle; wherein, the calibration trigger conditions include: the vehicle speed is within the range of 10km / h to 120km / h; the absolute value of the vehicle pitch angle is less than 0.2°;

[0017] The road surface is determined to be undulating based on the pitch angular velocity, pitch angle, and z-axis acceleration and displacement information output by the vehicle-mounted inertial navigation system.

[0018] Furthermore, the step of constructing a first change relationship corresponding to the first data, and determining and generating second data corresponding to the radar elevation angle based on the first change relationship, further includes:

[0019] A fourth set of data corresponding to the real-time operating status of the vehicle and within the range of the radar is generated and acquired. Based on the fourth set of data, a fifth set of data relative to the radar baseline is calculated and generated. The range of the radar is 3.5m to 35m. The fourth set of data is ground point cloud data. The fifth set of data is altitude data.

[0020] The system determines the trend of the fifth data point as the distance increases in real time. If the fifth data point increases with the distance, the radar is determined to be tilted upwards; if the fifth data point decreases with the distance, the radar is determined to be tilted downwards.

[0021] Furthermore, the step of constructing a first change relationship corresponding to the first data, and determining and generating second data corresponding to the radar elevation angle based on the first change relationship, further includes:

[0022] A sixth data point corresponding to the real-time operating status of the vehicle and within the radar range is generated and acquired. Based on the sixth data point, a seventh data point relative to the radar baseline is calculated and generated. The radar range is 10m to 150m. The sixth data point is high-altitude static target data with a height of not less than 3m. The seventh data point is altitude data. The lowest point of the target cloud cluster is taken as the reference point.

[0023] The system determines the trend of the seventh data as the distance decreases in real time. If the seventh data increases as the distance decreases, the radar is determined to be tilted upwards; if the seventh data decreases as the distance decreases, the radar is determined to be tilted downwards.

[0024] Furthermore, the step of iteratively compensating the radar elevation angle based on the second data until the first change relationship meets the preset calibration completion conditions also includes:

[0025] Angle compensation is performed using either a fixed step size or a dynamically adjusted step size strategy. The initial compensation step size is 0.1°, and the step size is dynamically adjusted based on the compensation effect. The cumulative absolute value of the compensated angle does not exceed 3°. The dynamically adjusted step size strategy includes: initially using a larger step size for compensation; reducing the compensation step size when the height change pattern after compensation reverses; maintaining or increasing the compensation step size when the height change pattern after compensation does not reverse; until precise compensation is achieved using the minimum step size.

[0026] When calculating the height value, a lookup table method is used to obtain the sine function value; wherein, the lookup table method pre-stores sine function values ​​in the range of -5° to +5° at intervals of 0.1°.

[0027] The second objective of this invention is achieved as follows: the system is used to implement the self-calibration processing method for elevation angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets, the system comprising:

[0028] The first data generation and acquisition unit is used to generate and acquire first data corresponding to the real-time operating status of the vehicle and the radar detection, and to analyze and process the first data; wherein, the first data includes ground point cloud data and high-altitude target point cloud data;

[0029] The data construction and determination generation unit is used to construct a first change relationship corresponding to the first data, and based on the first change relationship, determine and generate second data corresponding to the radar elevation angle; wherein, the second data is installation deviation direction data; and the first change relationship is altitude change law data.

[0030] The data iterative compensation processing unit is used to iteratively compensate the radar elevation angle according to the second data until the first change relationship meets the preset calibration completion conditions.

[0031] Furthermore, the first data generation and acquisition unit further includes:

[0032] The first generation module is used to generate and acquire third data corresponding to the real-time operating status of the vehicle; wherein, the third data is data on the road surface undulation status monitored during vehicle operation.

[0033] The first determination module is used to determine and generate corresponding first control data based on the third data, and to start the pitch angle self-calibration program based on the first control data; wherein, the first control data is the control data when the preset calibration trigger condition is met;

[0034] And / or, the data construction and determination generation unit further includes:

[0035] The second generation module is used to generate and acquire fourth data corresponding to the real-time operating status of the vehicle and within the range of the radar, and to calculate and generate fifth data relative to the radar baseline based on the fourth data; wherein, the range of the radar is 3.5m to 35m; the fourth data is ground point cloud data; and the fifth data is altitude value data;

[0036] The second determination module is used to determine the changing trend of the fifth data with increasing distance in real time. If the fifth data increases with increasing distance, it is determined that the radar is installed upward; if the fifth data decreases with increasing distance, it is determined that the radar is installed downward.

[0037] And / or, the data iteration compensation processing unit further includes:

[0038] The first processing module is used to perform angle compensation processing using a fixed step size or a dynamically adjusted step size strategy. The initial compensation step size is 0.1°, and the step size is dynamically adjusted based on the compensation effect. The cumulative absolute value of the compensated angle does not exceed 3°. The dynamically adjusted step size strategy includes: initially using a larger step size for compensation; reducing the compensation step size when the height change pattern after compensation reverses; maintaining or increasing the compensation step size when the height change pattern after compensation does not reverse; until precise compensation is achieved using the minimum step size.

[0039] The second processing module is used to obtain sine function values ​​by looking up a table when calculating the height value; wherein the lookup table method pre-stores sine function values ​​in the range of -5° to +5° at intervals of 0.1°.

[0040] Furthermore, the first data generation and acquisition unit further includes:

[0041] The first construction module is used to construct calibration trigger conditions corresponding to the real-time operating status of the vehicle; wherein, the calibration trigger conditions include: the vehicle speed is within the range of 10km / h to 120km / h; and the absolute value of the vehicle pitch angle is less than 0.2°;

[0042] The third judgment module is used to determine whether the road surface is undulating based on the pitch angular velocity, pitch angle, and z-axis acceleration and displacement information output by the vehicle inertial navigation system.

[0043] And / or, the data construction and determination generation unit further includes:

[0044] The third generation module is used to generate and acquire sixth data corresponding to the real-time operating status of the vehicle and within the range of the radar, and to calculate and generate seventh data relative to the radar baseline based on the sixth data; wherein, the range of the radar is 10m to 150m; the sixth data is high-altitude static target data with a height of not less than 3m; the seventh data is altitude value data; the lowest point of the target point cloud cluster is taken as the reference point;

[0045] The fourth determination module is used to determine the changing trend of the seventh data as the distance decreases in real time. If the seventh data increases as the distance decreases, it is determined that the radar is installed upward; if the seventh data decreases as the distance decreases, it is determined that the radar is installed downward.

[0046] The third objective of this invention is achieved as follows: it includes a processor, a memory, and a control program for a 4D imaging millimeter-wave radar elevation angle self-calibration processing platform based on ground and high-altitude targets; wherein the processor executes the control program for the 4D imaging millimeter-wave radar elevation angle self-calibration processing platform based on ground and high-altitude targets, the control program is stored in the memory, and the control program implements the 4D imaging millimeter-wave radar elevation angle self-calibration processing method based on ground and high-altitude targets.

[0047] This invention generates and acquires first data corresponding to the real-time operating status of a vehicle and radar detection through a method, and analyzes and processes the first data. The first data includes ground point cloud data and high-altitude target point cloud data. A first change relationship corresponding to the first data is constructed, and based on the first change relationship, second data corresponding to the radar elevation angle is generated. The second data is installation deviation direction data; the first change relationship is altitude change pattern data; based on the second data, the radar elevation angle is iteratively compensated until the first change relationship meets preset calibration completion conditions. The invention also includes a corresponding system and platform to solve the problem of inaccurate altitude measurement caused by changes in radar installation angle during the vehicle's lifespan. It utilizes common road surfaces and high-altitude static targets as references to achieve simple, reliable, and computationally-free self-calibration.

[0048] In other words, this invention innovatively combines the analysis of the height variation patterns of point clouds on flat roads and static target point clouds at high altitudes, achieving intelligent diagnosis and precise compensation for radar elevation installation deviations. This method does not rely on dedicated calibration fields or complex calculations; it can automatically complete calibration using only common scenarios encountered in daily driving, offering significant advantages in terms of wide applicability and low cost. Its core algorithm employs intuitive height order criteria and a dynamic step-size compensation strategy, combining logical simplicity, computational efficiency, and ease of engineering implementation. Simultaneously, it can quickly converge to the radar's own angular measurement accuracy limit (±0.1°), ensuring high calibration accuracy. Ultimately, it fundamentally solves the problem of height measurement distortion caused by changes in installation position throughout the vehicle's lifecycle, providing continuous and reliable target height and elevation angle information for Advanced Driver Assistance Systems (ADAS) and autonomous driving functions, greatly improving the safety and reliability of system decision-making. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1(a) is a schematic diagram of vehicle-mounted millimeter-wave radar for measuring elevation angle / height;

[0051] (b) is a schematic diagram of vehicle-mounted millimeter-wave radar measuring pitch angle / height. θ1 is the pitch angle of ground point P1 measured by the radar, and θ2 is the pitch angle of high point P2 measured by the radar.

[0052] Figure 2(a) is a schematic diagram of the pitch angle / height detection of road surface points when the radar is installed without bias, based on the pitch angle self-calibration processing method of 4D imaging millimeter-wave radar for ground and high-altitude targets according to the present invention.

[0053] (b) is a schematic diagram of the pitch angle / height detection of road surface points based on the radar view when the radar is installed without bias, according to the self-calibration processing method of 4D imaging millimeter-wave radar based on ground and high-altitude targets of the present invention.

[0054] Figure 3(a) is a schematic diagram of the pitch angle / height detection of a road surface point when the radar is offset upward by an angle, according to the self-calibration processing method of 4D imaging millimeter-wave radar based on ground and high-altitude targets of the present invention.

[0055] (b) is a schematic diagram of the pitch angle / height detection of road surface points based on the radar view when the radar is offset upward by an angle in the self-calibration processing method of 4D imaging millimeter-wave radar based on ground and high-altitude targets of the present invention.

[0056] Figure 4(a) is a schematic diagram of the pitch angle / height detection of a road surface point when the radar is offset downward by an angle, according to the self-calibration processing method of pitch angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets of the present invention.

[0057] (b) is a schematic diagram of the pitch angle / height detection of road surface points based on the radar view when the radar installation is offset downward by an angle in the self-calibration processing method of 4D imaging millimeter-wave radar based on ground and high-altitude targets of the present invention.

[0058] Figure 5 This is a schematic diagram of the radar's height / elevation angle measurement of the lowest point of the target point cloud when the radar is installed without offset, as the car drives towards the target below the target. This is a self-calibration processing method for the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets according to the present invention.

[0059] Figure 6This is a schematic diagram of the radar's height / elevation angle measurement of the lowest point of the target point cloud when the radar is installed downwards and the car is driving below the target at a high altitude, according to the self-calibration processing method of 4D imaging millimeter-wave radar based on ground and high-altitude targets of the present invention.

[0060] Figure 7 This is a schematic diagram of the radar's height / elevation angle measurement of the lowest point of the target point cloud when the radar is installed upwards and the car is driving below the target at a high altitude, according to the self-calibration processing method of 4D imaging millimeter-wave radar based on ground and high-altitude targets of the present invention.

[0061] Figure 8 This is a schematic diagram of the radar elevation calibration process using ground points as targets, which is part of the self-calibration processing method for elevation angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets according to the present invention.

[0062] Figure 9 This is a schematic diagram of the radar elevation calibration process using a high-altitude target point as the target, which is part of the self-calibration processing method for elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets according to the present invention.

[0063] Figure 10 This is a schematic diagram of the process steps of a 4D imaging millimeter-wave radar elevation angle self-calibration method based on ground and high-altitude targets according to the present invention.

[0064] Figure 11 This is a schematic diagram of the elevation angle self-calibration processing system architecture of a 4D imaging millimeter-wave radar based on ground and high-altitude targets according to the present invention.

[0065] Figure 12 This is a schematic diagram of the elevation angle self-calibration processing platform architecture for a 4D imaging millimeter-wave radar based on ground and high-altitude targets according to the present invention. Detailed Implementation

[0066] To facilitate a clearer understanding of the objectives, technical solutions, and advantages of this invention, the invention will be further described below in conjunction with the accompanying drawings and specific embodiments. Those skilled in the art can easily understand other advantages and effects of this invention from the content disclosed in this specification.

[0067] This invention can also be implemented or applied through other different specific examples, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of this invention.

[0068] It should be noted that if the embodiments of the present invention involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0069] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Secondly, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0070] Preferably, the 4D imaging millimeter-wave radar elevation angle self-calibration processing method based on ground and high-altitude targets of the present invention is applied in one or more terminals or servers. The terminal is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0071] The terminal can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal can interact with the customer via a keyboard, mouse, remote control, touchpad, or voice control device.

[0072] This invention provides a method, system, and platform for self-calibration of elevation angles in 4D imaging millimeter-wave radar based on ground and high-altitude targets.

[0073] like Figure 10 The diagram shown is a flowchart of the pitch angle self-calibration processing method for 4D imaging millimeter-wave radar based on ground and high-altitude targets provided in an embodiment of the present invention.

[0074] In this embodiment, the 4D imaging millimeter-wave radar elevation angle self-calibration processing method based on ground and high-altitude targets can be applied to terminals or fixed terminals with display functions. The terminals are not limited to personal computers, smartphones, tablets, desktop computers or all-in-one computers with cameras, etc.

[0075] The self-calibration processing method for elevation angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets can also be applied to a hardware environment consisting of a terminal and a server connected to the terminal via a network. The network includes, but is not limited to, wide area networks (WANs), metropolitan area networks (MANs), or local area networks (LANs). The self-calibration processing method for elevation angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets in this embodiment of the invention can be executed by the server, by the terminal, or by both the server and the terminal.

[0076] For example, for a terminal requiring 4D imaging millimeter-wave radar elevation angle self-calibration processing based on ground and high-altitude targets, the elevation angle self-calibration processing function provided by the method of this invention can be directly integrated into the terminal, or a client for implementing the method of this invention can be installed. Alternatively, the method provided by this invention can also run on servers or other devices in the form of a Software Development Kit (SDK). The SDK provides an interface for the 4D imaging millimeter-wave radar elevation angle self-calibration processing function based on ground and high-altitude targets, allowing the terminal or other devices to implement the elevation angle self-calibration processing function through the provided interface. The invention will be further described below with reference to the accompanying drawings.

[0077] like Figure 10 As shown, this invention provides a method for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets.

[0078] S1. Generate and acquire first data corresponding to the real-time operating status of the vehicle and radar detection, and analyze and process the first data; wherein, the first data includes ground point cloud data and high-altitude target point cloud data;

[0079] S2. Construct a first change relationship corresponding to the first data, and based on the first change relationship, determine and generate second data corresponding to the radar elevation angle; wherein, the second data is installation deviation direction data; the first change relationship is altitude change law data;

[0080] S3. Based on the second data, iteratively compensate the radar elevation angle until the first change relationship meets the preset calibration completion conditions.

[0081] The process of generating and acquiring first data corresponding to the real-time vehicle status and radar detection, and analyzing and processing the first data, further includes:

[0082] S11. Generate and acquire third data corresponding to the real-time operating status of the vehicle; wherein, the third data is road surface undulation data monitored during vehicle operation;

[0083] S12. Based on the third data, determine and generate corresponding first control data, and start the pitch angle self-calibration program based on the first control data; wherein, the first control data is the control data when the preset calibration trigger condition is met.

[0084] The process of generating and acquiring first data corresponding to the real-time vehicle status and radar detection, and analyzing and processing the first data, further includes:

[0085] S13. Construct calibration trigger conditions corresponding to the real-time operating status of the vehicle; wherein, the calibration trigger conditions include: the vehicle speed is within the range of 10km / h to 120km / h; the absolute value of the vehicle pitch angle is less than 0.2°;

[0086] S14. Based on the pitch angular velocity, pitch angle, and z-axis acceleration and displacement information output by the vehicle-mounted inertial navigation system, it is determined that the road surface is undulating.

[0087] The step of constructing a first change relationship corresponding to the first data, and determining and generating second data corresponding to the radar elevation angle based on the first change relationship, further includes:

[0088] S21. Generate and acquire fourth data corresponding to the real-time operating status of the vehicle and within the range of the radar, and calculate and generate fifth data relative to the radar baseline based on the fourth data; wherein, the range of the radar is 3.5m to 35m; the fourth data is ground point cloud data; the fifth data is altitude value data;

[0089] S22. In real time, determine the trend of the fifth data as the distance increases. If the fifth data increases as the distance increases, determine that the radar is installed upward; if the fifth data decreases as the distance increases, determine that the radar is installed downward.

[0090] The step of constructing a first change relationship corresponding to the first data, and determining and generating second data corresponding to the radar elevation angle based on the first change relationship, further includes:

[0091] S23. Generate and acquire sixth data corresponding to the real-time operating status of the vehicle and within the range of the radar, and calculate and generate seventh data relative to the radar baseline based on the sixth data; wherein, the range of the radar is 10m to 150m; the sixth data is high-altitude static target data with a height of not less than 3m; the seventh data is altitude value data; take the lowest point of the target cloud cluster as the reference point;

[0092] S24. Real-time determination of the changing trend of the seventh data as the distance decreases. If the seventh data increases as the distance decreases, it is determined that the radar is installed upwards; if the seventh data decreases as the distance decreases, it is determined that the radar is installed downwards.

[0093] The step of iteratively compensating the radar elevation angle based on the second data until the first change relationship meets the preset calibration completion conditions further includes:

[0094] S31. Angle compensation is performed using a fixed step size or a dynamically adjusted step size strategy. The initial compensation step size is 0.1°, and the step size is dynamically adjusted according to the compensation effect. The cumulative absolute value of the compensation angle does not exceed 3°. The dynamic step size adjustment strategy includes: initially using a larger step size for compensation; reducing the compensation step size when the height change pattern after compensation reverses; maintaining or increasing the compensation step size when the height change pattern after compensation does not reverse; until precise compensation is completed using the minimum step size.

[0095] S32. When calculating the height value, a lookup table method is used to obtain the sine function value; wherein, the lookup table method pre-stores sine function values ​​in the range of -5° to +5° at intervals of 0.1°.

[0096] Specifically, in this embodiment of the invention, the 4D imaging radar pitch self-calibration requires the vehicle to be traveling on a road with no undulations. The system uses pitch angle information (pitch angular velocity, pitch angle) and acceleration and displacement information along the z-axis output by the onboard inertial navigation IMU system on the domain controller to determine whether the road surface meets the "no undulation" calibration condition. The radar measures the height of ground points at different distances and statistically analyzes the height of high-altitude targets such as gantries, road signs, and traffic lights. Based on these height variation patterns, calibration actions are performed, continuously adjusting the compensation value towards a set target until pitch angle calibration is completed.

[0097] The vehicle is traveling or stationary on a flat road. Assuming the 4D imaging radar R is installed directly in front of the vehicle, typically at a height of 30-50cm above the ground, the radar measures road surface points P1, P2, ..., creating a point cloud of scattering points. In Figure 2(a), the radar uses the formula H=r*sinθ to measure the heights of each road surface point relative to the radar baseline, which are P1G1, P2G2, P3G3, P4G4, etc., and are all equal. Subtracting the radar installation height, the height of these points relative to the ground is 0. Radar angle measurement always uses the antenna surface normal as the baseline, i.e., the dashed line in Figure 2. When the radar is installed without bias, the baseline is parallel to the ground. The road surface points seen from the radar's perspective are shown in Figure 2(b).

[0098] Figure 3 is a schematic diagram of the detection of road surface points when the radar is installed at a small upward tilt. In Figure 3(a), the heights of each road surface point relative to the radar baseline measured by the radar using the formula H=r*sinθ are P1G1, P2G2, P3G3, P4G4, etc., and they gradually increase in height as they are farther away from the radar. After subtracting the radar installation height, the "ground" formed by these points is shown in Figure 3(b), which shows a downward slope. This is because the radar is unaware of its own installation offset and still assumes that the baseline is parallel to the horizontal plane.

[0099] Figure 4 is a schematic diagram of the detection of road surface points when the radar is installed at a small downward angle. In Figure 4(a), the height of each point on the road surface relative to the radar baseline gradually decreases as the distance from the radar increases. After subtracting the radar installation height, the "ground" fitted by these points is shown in Figure 4(b), which shows an upward slope.

[0100] Calculating the slope relationship between H and r is not necessary for pitch calibration.

[0101] The quantitative description is as follows: For road surface points detected by radar, points P1, P2, ..., P10 can be selected from near to far, requiring the distance between adjacent points to be greater than 5m. Calculate the "height" H1 to H10 of P1 to P10 relative to the radar baseline. If, except for at most two points (excluding radar-detected anomalies), the remaining points satisfy the order H1 < H2 < ... < H10, then the radar installation angle is tilted upwards. Conversely, the radar installation angle is tilted downwards.

[0102] Besides using flat roads as targets, high-altitude, strongly scattering targets can also be chosen as targets for pitch angle calibration, such as traffic lights, metal gantries, and metal road signs. To address the issue of point clouds of highly scattering objects at different altitudes, the lowest point of the point cloud cluster can be used as the reference point for the target.

[0103] The point cloud of the lowest point of a high-altitude, strongly scattering target, measured by a radar mounted on a vehicle at times t1, t2, and t3, is shown in the figure above. The vehicle is traveling from a distance towards the area below the high-altitude target. When the radar is mounted without tilt, the lowest point of the target's point cloud is at the same height at times t1, t2, and t3. The range threshold for the high-altitude, strongly scattering target is: 10m ≤ r ≤ 150m. High-altitude static strongly scattering targets ("static strongly scattering targets" refers to targets that can form stable static track points in the radar data processing algorithm) at this range can meet the signal-to-noise ratio requirements.

[0104] For indicators such as signal-to-noise ratio, density and stability of point cloud clusters, and RCS, the static high-altitude target tracks and their point clouds with stable ID numbers output by the clustering algorithm of radar data processing can be selected.

[0105] The crossbeams of the gantry and road signs, as long as they meet the above conditions, can also be used as targets for elevation calibration. Vehicles ahead are usually moving targets, and their point clouds contain speed information, making them easy to distinguish from static targets such as road signs. Even if there are stationary, tall vehicles ahead, their point cloud distribution will vary in height, failing to meet the requirement that "the lowest point cloud height of the target is ≥3m".

[0106] When the radar is installed at a downward tilt, the heights of the lowest point of the target point cloud relative to the radar baseline measured at times t1, t2, and t3 are SG1, SG2, and SG3, respectively. The closer the radar is to the target, the smaller the height measured by the radar.

[0107] When the radar is installed at an upward tilt, the heights of the lowest point of the target point cloud relative to the radar baseline measured at times t1, t2, and t3 are SG1, SG2, and SG3, respectively. The closer the radar is to the target, the higher the height measured by the radar.

[0108] In summary, when the radar elevation plane is tilted, the absolute changes in the height of each point cloud on the road surface relative to the radar baseline detected by the radar can be summarized by two rules, A and B:

[0109] A. When tilting upwards, the farther away from the radar, the greater the absolute value of the altitude;

[0110] B. When tilting downwards, the further away from the radar, the smaller the absolute value of the altitude;

[0111] When the radar elevation plane is tilted, the absolute change in the height of the lowest point of the point cloud of high-altitude targets detected by the radar relative to the radar baseline can be summarized by two rules: C and D. The lowest point is taken from the statistically lowest percentile of the point cloud. These points typically correspond to the lower left and right corners of road signs, because the RCS of the corners of metallic reflective objects is very high. The lowest point of the point cloud of a specific high-altitude static target will fluctuate in each frame, but the fluctuation range is at most tens of centimeters at a distance, having little impact on the elevation angle estimation. At closer distances, the point cloud becomes more stable as the vehicle gets closer to the road sign (because the signal-to-noise ratio increases).

[0112] C. When tilting upwards, the absolute value of the altitude increases the further away from the radar.

[0113] D. When tilting downwards, the further away from the radar, the smaller the absolute value of the altitude;

[0114] Specifically, calculating the slope k is equivalent to calculating the calibration compensation angle. This is rather cumbersome. One of the advantages of this patent is that it is "simple and easy to implement, does not involve complex mathematical derivations, and does not require direct calculation of the calibration compensation angle".

[0115] The following can be a more quantitative description of (A / B / C / D): A. For road surface points detected by radar, N points, P1, P2, ..., PN, are selected from near to far, requiring a distance > 5m between adjacent points. The relative "height" H1 to HN of P1 to PN from the radar baseline is calculated. If, except for at most m points, the remaining points satisfy the order H1 < H2 < ... < HN, then the radar installation angle is tilted upwards. B. Conversely, for road surface points detected by radar, if, except for at most m points, the remaining points satisfy the order H1 > H2 > ... > HN, then the radar installation angle is tilted downwards. C. For the absolute value of the height of the lowest point of the high-altitude target point cloud relative to the radar baseline, as the vehicle gradually approaches the target, the radar measures the lowest point of the target point cloud, P1, P2, ..., PN, etc., from far to near, requiring a distance > 5m between adjacent points. Calculate the relative "height" H1 to HN of P1 to PN from the radar baseline. If, except for at most m points, the remaining points satisfy the order H1 < H2 < ... < HN, then the radar installation angle is tilted upwards. Conversely, for the lowest point of the high-altitude target cloud detected by the radar, if, except for at most m points, the remaining points satisfy the order H1 > H2 > ... > HN, then the radar installation angle is tilted downwards. N can be chosen from 10 to 20, and m can be chosen from 2.

[0116] When the radar is installed without pitch or tilt, the height of the lowest point cloud of the road surface and the lowest point cloud of high-altitude targets relative to the radar baseline will not change with their distance from the radar.

[0117] The elevation angle measurement accuracy of a 4D imaging radar is approximately ±0.1° to ±0.2°. The area with higher elevation angle measurement accuracy is within ±5°, approaching ±0.1°. Therefore, when calibrating elevation, using point clouds within a ±5° range can improve calibration accuracy. For ground point clouds, assuming a radar installation height of 0.3m, the distance r corresponding to an elevation angle of -5° is 3.44m; the distance r corresponding to an elevation angle of -0.5° is 34.38m. For high-altitude, strongly scattering targets, the typical altitude is 4m. The distance r corresponding to an elevation angle of +5° is 42.45m; the distance r corresponding to an elevation angle of -0.5° is 424m (in practice, the furthest 4D imaging radar measures high-altitude targets at similar altitudes is 200m). Calibration can be performed by directly selecting road points and high-altitude target point clouds within the elevation angle range of (-5°, -0.5°) and (0.5°, 5°) measured by the radar.

[0118] The pitch self-calibration algorithm flow is designed as follows:

[0119] 1. When the vehicle's radar is traveling on a long stretch of road without undulations, the radar pitch self-calibration program is initiated. In this process, the pitch angle information (pitch angular velocity, pitch angle) and the acceleration and displacement information of the z-axis output by the vehicle's onboard inertial navigation IMU system on the domain controller are used to determine whether the road surface meets the "undulation" calibration conditions.

[0120] The vehicle's speed is v0, and the self-calibration trigger condition is:

[0121] 10km / h≤v0≤120km / h;

[0122] During the driving phase, the absolute value of the pitch angle measured by the vehicle is <0.2°.

[0123] If the road conditions suddenly become unsuitable during the calibration process, the calibration process will be stopped and the current data will be discarded.

[0124] Under the above conditions, most static points with negative pitch angles are ground points. Of course, outliers need to be filtered out, which can be done based on the height and distance information of the candidate point cloud. The algorithm includes a ground point filtering module.

[0125] 2. Radar statistical analysis of ground point cloud elevation angle information from near to far: Statistical analysis of the distance *r* and elevation angle *α* of the ground point cloud measured by the radar, where *r* ranges from 3.5m to 35m. The absolute value of the point cloud's height above the radar baseline is calculated using the formula H = *r*sinα and stored.

[0126] 3. When the absolute value of the height of the ground point cloud from the radar baseline changes with distance r in accordance with A, compensate the radar elevation angle by -0.1°—that is, subtract 0.1° from the elevation angle of the statistically analyzed road point cloud. Then recalculate the height value using the formula H=r*sin(α'), where α' is the compensated elevation angle.

[0127] (1) If the ground point cloud still conforms to A, continue to compensate the elevation angle by -0.1°. The program ends when the ground point changes to conform to B or the cumulative compensation angle value is <-3°. If the cumulative compensation angle sum is ≥-3° and the change from A to B is achieved, the compensation stops and proceeds to step 4 (optional), or the calibration is completed and the program exits. If the cumulative compensation angle value is <-3° and the change to conform to B has not yet occurred, it indicates that the radar installation deviation angle exceeds the calibrable range. The radar angle deviation is too large to be restored to normal by self-calibration. The radar stops working and reports a fault code to the domain controller.

[0128] (2) When the absolute value of the height of the ground point cloud from the radar baseline changes with distance r in accordance with B, compensate the radar elevation angle by +0.1°—that is, add 0.1° to the elevation angle of the statistical road point cloud. Then recalculate the height value using the formula H=r*sin(α'), where α' is the compensated elevation angle. Continue to compensate the elevation angle by +0.1° when the ground point cloud still conforms to B. The program ends when the ground point changes to conform to A or the cumulative compensated angle value is greater than 3°. If the cumulative compensated angle is less than 3° and the change from B to A is achieved, stop compensation and proceed to step 4 (optional), or complete calibration and exit. If the cumulative compensated angle value is greater than 3° and the change has not yet conformed to A, it indicates that the radar installation deviation angle exceeds the calibrable range, the radar angle deviation is too large to be restored to normal through self-calibration, the radar stops working and reports a fault code to the domain controller.

[0129] 4. The above procedure completes the ground point calibration of the radar elevation angle. Next is the optional calibration using a high-altitude target as the target. When the car is driving on a long stretch of flat road without undulations, and a stable high-altitude target is detected between 150m and 200m and it is approaching it, the radar counts the elevation angle α and distance r of the lowest point cloud of the high-altitude target from near to far, and calculates and stores the absolute value of the height of the point cloud from the radar baseline using the formula H=r*sinα: (1) When the absolute value of the height of the lowest point cloud of the high-altitude target from the radar baseline changes with distance r in accordance with C, the radar elevation angle is compensated by +0.1° - that is, 0.1° is added to the statistical elevation angle of the lowest point cloud of the high-altitude target. Then the height value is recalculated using the formula H=r*sin(α'), where α' is the compensated elevation angle. When the change in point cloud height still conforms to C, the elevation angle is compensated by +0.1°. The procedure ends when the change conforms to D; (2) When the absolute value of the height of the lowest point cloud of the high-altitude target relative to the radar baseline conforms to D, the compensation steps opposite to (1) are executed.

[0130] 5. The joint calibration of the above-mentioned road surface and high-altitude targets can be carried out several times. Multiple road sections and several different high-altitude targets can be selected for calibration to make the results more stable.

[0131] The 0.1° compensation step and minimum interval were chosen because the angular accuracy of the 4D imaging radar is approximately ±0.1°. Smaller steps would exceed the radar's accuracy range and would be meaningless.

[0132] The method described above, which compensates for angles at 0.1° intervals, can be improved to a dynamic step-size adjustment strategy (details of which can be found in the corresponding Python code). Initially, a larger step size is used (0.1° to 1°, with a minimum interval of 0.1°). The step size is increased upon success and decreased upon failure, quickly approaching the target compensation value. For example, initially compensate by 1°. If a flip is achieved, the compensation value is halved to 0.5°. If a 0.5° compensation still results in a flip, the compensation value is halved again and rounded down to 0.3°, and so on until 0.1° is reached. If the flip is not achieved, continue compensating by 1°. This iterative approach accelerates convergence.

[0133] The Python code for dynamically adjusting the step size using road surface points as calibration targets is as follows (similar for aerial targets):

[0134] import math

[0135] def find_compensation_angle(alpha,condition_A,condition_B):

[0136] """

[0137] Implementation of variable step size compensation algorithm

[0138] parameter:

[0139] alpha: Initial angle (degrees), i.e., α

[0140] condition_A: The function to evaluate condition A. It takes the value of sinα as input and returns whether the condition is satisfied.

[0141] condition_B: The function to evaluate condition B. It takes a sinα value as input and returns whether the condition is satisfied.

[0142] return:

[0143] Final compensation angle (degrees) or None (calibration failed)

[0144] """

[0145] step_sequence=[1.0, 0.5, 0.3, 0.2, 0.1] #Step size change sequence

[0146] current_step_index=0

[0147] compensation=0.0

[0148] #Initial state determination

[0149] sin_val = math.sin(math.radians(alpha)) # sin_val is the value of sin(α)

[0150] initial_in_A=condition_A(sin_val)

[0151] initial_in_B=condition_B(sin_val)

[0152] if initial_in_A==initial_in_B:

[0153] raise ValueError("The initial state cannot simultaneously satisfy or fail to satisfy conditions A and B")

[0154] #Determine the compensation direction: positive when B is satisfied, negative when A is satisfied.

[0155] direction=1 if initial_in_B else -1

[0156] while current_step_index <len(step_sequence):

[0157] current_step=step_sequence[current_step_index]*direction

[0158] #Check cumulative compensation value

[0159] if abs(compensation+current_step)>3.0:

[0160] print("Calibration failed: Cumulative compensation value exceeds 3° limit")

[0161] return None

[0162] # Attempt to compensate

[0163] temp_compensation=compensation+current_step

[0164] temp_alpha=alpha+temp_compensation

[0165] sin_va=math.sin(math.radians(temp_alpha))

[0166] new_in_A = condition_A(sin_val)

[0167] new_in_B = condition_B(sin_val)

[0168] #Check if a flip has occurred

[0169] flip_occurred=(initial_in_A and new_in_B)or(initial_in_B and new_in_A)

[0170] if not flip_occurred:

[0171] #No flip, continue with current step size compensation

[0172] compensation=temp_compensation

[0173] else:

[0174] if current_step_index <len(step_sequence)-1:

[0175] #If a flip occurs, decrease the step size.

[0176] current_step_index += 1

[0177] else:

[0178] #This is already the minimum step size, return the result.

[0179] return compensation

[0180] return compensation

[0181] Calculating altitude information involves calculating a sine function, which can be improved by using a lookup table to speed up the calculation: Pre-store the sine function values ​​within the angle range used for elevation angle calibration, such as sin(-5°), sin(-4.9°), ... up to sin(5°), in 0.1° intervals, because the radar angle measurement accuracy is 0.1°. Calculate H = r * sinα by directly looking up the value in the table.

[0182] To achieve the above objectives, the present invention also provides a 4D imaging millimeter-wave radar elevation angle self-calibration processing system based on ground and high-altitude targets, such as... Figure 11 As shown, the system is used to implement the self-calibration processing method for elevation angle of 4D imaging millimeter-wave radar based on ground and high-altitude targets. The system includes:

[0183] The first data generation and acquisition unit is used to generate and acquire first data corresponding to the real-time operating status of the vehicle and the radar detection, and to analyze and process the first data; wherein, the first data includes ground point cloud data and high-altitude target point cloud data;

[0184] The data construction and determination generation unit is used to construct a first change relationship corresponding to the first data, and based on the first change relationship, determine and generate second data corresponding to the radar elevation angle; wherein, the second data is installation deviation direction data; and the first change relationship is altitude change law data.

[0185] The data iterative compensation processing unit is used to iteratively compensate the radar elevation angle according to the second data until the first change relationship meets the preset calibration completion conditions.

[0186] The first data generation and acquisition unit further includes:

[0187] The first generation module is used to generate and acquire third data corresponding to the real-time operating status of the vehicle; wherein, the third data is data on the road surface undulation status monitored during vehicle operation.

[0188] The first determination module is used to determine and generate corresponding first control data based on the third data, and to start the pitch angle self-calibration program based on the first control data; wherein, the first control data is the control data when the preset calibration trigger condition is met;

[0189] And / or, the data construction and determination generation unit further includes:

[0190] The second generation module is used to generate and acquire fourth data corresponding to the real-time operating status of the vehicle and within the range of the radar, and to calculate and generate fifth data relative to the radar baseline based on the fourth data; wherein, the range of the radar is 3.5m to 35m; the fourth data is ground point cloud data; and the fifth data is altitude value data;

[0191] The second determination module is used to determine the changing trend of the fifth data with increasing distance in real time. If the fifth data increases with increasing distance, it is determined that the radar is installed upward; if the fifth data decreases with increasing distance, it is determined that the radar is installed downward.

[0192] And / or, the data iteration compensation processing unit further includes:

[0193] The first processing module is used to perform angle compensation processing using a fixed step size or a dynamically adjusted step size strategy. The initial compensation step size is 0.1°, and the step size is dynamically adjusted based on the compensation effect. The cumulative absolute value of the compensated angle does not exceed 3°. The dynamically adjusted step size strategy includes: initially using a larger step size for compensation; reducing the compensation step size when the height change pattern after compensation reverses; maintaining or increasing the compensation step size when the height change pattern after compensation does not reverse; until precise compensation is achieved using the minimum step size.

[0194] The second processing module is used to obtain sine function values ​​by looking up a table when calculating the height value; wherein the lookup table method pre-stores sine function values ​​in the range of -5° to +5° at intervals of 0.1°.

[0195] The first data generation and acquisition unit further includes:

[0196] The first construction module is used to construct calibration trigger conditions corresponding to the real-time operating status of the vehicle; wherein, the calibration trigger conditions include: the vehicle speed is within the range of 10km / h to 120km / h; and the absolute value of the vehicle pitch angle is less than 0.2°;

[0197] The third judgment module is used to determine whether the road surface is undulating based on the pitch angular velocity, pitch angle, and z-axis acceleration and displacement information output by the vehicle inertial navigation system.

[0198] And / or, the data construction and determination generation unit further includes:

[0199] The third generation module is used to generate and acquire sixth data corresponding to the real-time operating status of the vehicle and within the range of the radar, and to calculate and generate seventh data relative to the radar baseline based on the sixth data; wherein, the range of the radar is 10m to 150m; the sixth data is high-altitude static target data with a height of not less than 3m; the seventh data is altitude value data; the lowest point of the target point cloud cluster is taken as the reference point;

[0200] The fourth determination module is used to determine the changing trend of the seventh data as the distance decreases in real time. If the seventh data increases as the distance decreases, it is determined that the radar is installed upward; if the seventh data decreases as the distance decreases, it is determined that the radar is installed downward.

[0201] In the system embodiment of the present invention, the specific details of the method steps involved in the pitch angle self-calibration processing of a 4D imaging millimeter-wave radar based on ground and high-altitude targets have been described above. That is to say, the functional components in the system are used to implement the steps or sub-steps in the above method embodiment, and will not be repeated here.

[0202] To achieve the above objectives, the present invention also provides a 4D imaging millimeter-wave radar elevation angle self-calibration processing platform based on ground and high-altitude targets, such as... Figure 12 As shown, the system includes a processor, a memory, and a control program for a 4D imaging millimeter-wave radar elevation angle self-calibration processing platform based on ground and high-altitude targets. The processor executes the control program, which is stored in the memory. This control program implements the steps of the 4D imaging millimeter-wave radar elevation angle self-calibration processing method based on ground and high-altitude targets. For example: S1. Generate and acquire first data corresponding to the real-time operating status of the vehicle and the radar detection, and analyze and process the first data; wherein, the first data includes ground point cloud data and high-altitude target point cloud data; S2. Construct a first change relationship corresponding to the first data, and based on the first change relationship, determine and generate second data corresponding to the radar pitch angle; wherein, the second data is installation deviation direction data; the first change relationship is altitude change law data; S3. According to the second data, iteratively compensate and process the radar pitch angle until the first change relationship meets the preset calibration completion conditions.

[0203] The specific details of the steps have been explained above and will not be repeated here.

[0204] In this embodiment of the invention, the built-in processor of the 4D imaging millimeter-wave radar elevation angle self-calibration processing platform based on ground and high-altitude targets can be composed of integrated circuits. For example, it can be composed of a single packaged integrated circuit, or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor connects to various components using various interfaces and lines, and executes programs or units stored in memory, as well as calling data stored in memory, to perform various functions and process data for the 4D imaging millimeter-wave radar elevation angle self-calibration based on ground and high-altitude targets.

[0205] The memory is used to store program code and various data. It is installed in the pitch angle self-calibration processing platform of the 4D imaging millimeter-wave radar based on ground and high-altitude targets, and can complete the storage and retrieval of programs or data at high speed and automatically during operation.

[0206] The memory includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0207] This invention generates and acquires first data corresponding to the real-time operating status of a vehicle and radar detection through a method, and analyzes and processes the first data. The first data includes ground point cloud data and high-altitude target point cloud data. A first change relationship corresponding to the first data is constructed, and based on the first change relationship, second data corresponding to the radar elevation angle is generated. The second data is installation deviation direction data; the first change relationship is altitude change pattern data; based on the second data, the radar elevation angle is iteratively compensated until the first change relationship meets preset calibration completion conditions. The invention also includes a corresponding system and platform to solve the problem of inaccurate altitude measurement caused by changes in radar installation angle during the vehicle's lifespan. It utilizes common road surfaces and high-altitude static targets as references to achieve simple, reliable, and computationally-free self-calibration.

[0208] In other words, this invention innovatively combines the analysis of the height variation patterns of point clouds on flat roads and static target point clouds at high altitudes, achieving intelligent diagnosis and precise compensation for radar elevation installation deviations. This method does not rely on dedicated calibration fields or complex calculations; it can automatically complete calibration using only common scenarios encountered in daily driving, offering significant advantages in terms of wide applicability and low cost. Its core algorithm employs intuitive height order criteria and a dynamic step-size compensation strategy, combining logical simplicity, computational efficiency, and ease of engineering implementation. Simultaneously, it can quickly converge to the radar's own angular measurement accuracy limit (±0.1°), ensuring high calibration accuracy. Ultimately, it fundamentally solves the problem of height measurement distortion caused by changes in installation position throughout the vehicle's lifecycle, providing continuous and reliable target height and elevation angle information for Advanced Driver Assistance Systems (ADAS) and autonomous driving functions, greatly improving the safety and reliability of system decision-making.

[0209] In other words, the present invention is simple and easy to implement, does not involve complex mathematical derivations, and does not require direct calculation of the calibration compensation angle; moreover, it has a high-speed convergence method and table lookup to accelerate calculation; it has low requirements for the calibration scene and wide adaptability: flat roads are the most common, and high-altitude objects with strong scattering are also common. Furthermore, this solution does not limit the specific height of the high-altitude object (≥3m is sufficient). When calibrating the pitch angle using the road surface as a reference, both stationary and moving objects are acceptable. When using a high-altitude object as a reference, the vehicle must be in motion.

[0210] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets, characterized in that, The method includes the following steps: Generate and acquire first data corresponding to the real-time operating status of the vehicle and radar detection, and analyze and process the first data; wherein, the first data includes ground point cloud data and high-altitude target point cloud data; A first change relationship corresponding to the first data is constructed, and based on the first change relationship, second data corresponding to the radar elevation angle is determined and generated; wherein, the second data is installation deviation direction data; and the first change relationship is altitude change pattern data. Based on the second data, the radar elevation angle is iteratively compensated until the first change relationship meets the preset calibration completion conditions.

2. The method for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets according to claim 1, characterized in that, The process of generating and acquiring first data corresponding to the real-time vehicle status and radar detection, and analyzing and processing the first data, further includes: Generate and acquire third data corresponding to the real-time operating status of the vehicle; wherein, the third data is data on the road surface undulation status monitored during vehicle operation; Based on the third data, corresponding first control data is generated, and the pitch angle self-calibration program is started based on the first control data; wherein, the first control data is the control data when the preset calibration trigger condition is met.

3. A method for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets, as described in claim 1 or 2, characterized in that... The process of generating and acquiring first data corresponding to the real-time vehicle status and radar detection, and analyzing and processing the first data, further includes: Establish calibration trigger conditions corresponding to the real-time operating status of the vehicle; wherein, the calibration trigger conditions include: the vehicle speed is within the range of 10km / h to 120km / h; the absolute value of the vehicle pitch angle is less than 0.2°; The road surface is determined to be undulating based on the pitch angular velocity, pitch angle, and z-axis acceleration and displacement information output by the vehicle-mounted inertial navigation system.

4. The method for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets according to claim 1, characterized in that, The step of constructing a first change relationship corresponding to the first data, and determining and generating second data corresponding to the radar elevation angle based on the first change relationship, further includes: A fourth set of data corresponding to the real-time operating status of the vehicle and within the range of the radar is generated and acquired. Based on the fourth set of data, a fifth set of data relative to the radar baseline is calculated and generated. The range of the radar is 3.5m to 35m. The fourth set of data is ground point cloud data. The fifth set of data is altitude data. The system determines the trend of the fifth data point as the distance increases in real time. If the fifth data point increases with the distance, the radar is determined to be tilted upwards; if the fifth data point decreases with the distance, the radar is determined to be tilted downwards.

5. A method for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets, as described in claim 1 or 4, characterized in that... The step of constructing a first change relationship corresponding to the first data, and determining and generating second data corresponding to the radar elevation angle based on the first change relationship, further includes: A sixth data point corresponding to the real-time operating status of the vehicle and within the radar range is generated and acquired. Based on the sixth data point, a seventh data point relative to the radar baseline is calculated and generated. The radar range is 10m to 150m. The sixth data point is high-altitude static target data with a height of not less than 3m. The seventh data point is altitude data. The lowest point of the target cloud cluster is taken as the reference point. The system determines the trend of the seventh data as the distance decreases in real time. If the seventh data increases as the distance decreases, the radar is determined to be tilted upwards; if the seventh data decreases as the distance decreases, the radar is determined to be tilted downwards.

6. The method for self-calibrating the elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets according to claim 1, characterized in that, The step of iteratively compensating the radar elevation angle based on the second data until the first change relationship meets the preset calibration completion conditions further includes: Angle compensation is performed using either a fixed step size or a dynamically adjusted step size strategy. The initial compensation step size is 0.1°, and the step size is dynamically adjusted based on the compensation effect. The cumulative absolute value of the compensated angle does not exceed 3°. The dynamically adjusted step size strategy includes: initially using a larger step size for compensation; reducing the compensation step size when the height change pattern after compensation reverses; maintaining or increasing the compensation step size when the height change pattern after compensation does not reverse; until precise compensation is achieved using the minimum step size. When calculating the height value, a lookup table method is used to obtain the sine function value; wherein, the lookup table method pre-stores sine function values ​​in the range of -5° to +5° at intervals of 0.1°.

7. A self-calibration processing system for elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets, characterized in that, The system is used to implement the 4D imaging millimeter-wave radar elevation angle self-calibration processing method based on ground and high-altitude targets as described in any one of claims 1 to 6, and the system includes: The first data generation and acquisition unit is used to generate and acquire first data corresponding to the real-time operating status of the vehicle and the radar detection, and to analyze and process the first data; wherein, the first data includes ground point cloud data and high-altitude target point cloud data; The data construction and determination generation unit is used to construct a first change relationship corresponding to the first data, and based on the first change relationship, determine and generate second data corresponding to the radar elevation angle; wherein, the second data is installation deviation direction data; and the first change relationship is altitude change law data. The data iterative compensation processing unit is used to iteratively compensate the radar elevation angle according to the second data until the first change relationship meets the preset calibration completion conditions.

8. The 4D imaging millimeter-wave radar elevation angle self-calibration processing system based on ground and high-altitude targets according to claim 7, characterized in that, The first data generation and acquisition unit further includes: The first generation module is used to generate and acquire third data corresponding to the real-time operating status of the vehicle; wherein, the third data is data on the road surface undulation status monitored during vehicle operation. The first determination module is used to determine and generate corresponding first control data based on the third data, and to start the pitch angle self-calibration program based on the first control data; wherein, the first control data is the control data when the preset calibration trigger condition is met; And / or, the data construction and determination generation unit further includes: The second generation module is used to generate and acquire fourth data corresponding to the real-time operating status of the vehicle and within the range of the radar, and to calculate and generate fifth data relative to the radar baseline based on the fourth data; wherein, the range of the radar is 3.5m to 35m; the fourth data is ground point cloud data; and the fifth data is altitude value data; The second determination module is used to determine the trend of the fifth data with increasing distance in real time. If the fifth data increases with increasing distance, it is determined that the radar is installed upward; if the fifth data decreases with increasing distance, it is determined that the radar is installed downward. And / or, the data iteration compensation processing unit further includes: The first processing module is used to perform angle compensation processing using a fixed step size or a dynamically adjusted step size strategy. The initial compensation step size is 0.1°, and the step size is dynamically adjusted based on the compensation effect. The cumulative absolute value of the compensated angle does not exceed 3°. The dynamically adjusted step size strategy includes: initially using a larger step size for compensation; reducing the compensation step size when the height change pattern after compensation reverses; maintaining or increasing the compensation step size when the height change pattern after compensation does not reverse; until precise compensation is achieved using the minimum step size. The second processing module is used to obtain sine function values ​​by looking up a table when calculating the height value; wherein the lookup table method pre-stores sine function values ​​in the range of -5° to +5° at intervals of 0.1°.

9. A 4D imaging millimeter-wave radar elevation angle self-calibration processing system based on ground and high-altitude targets according to claim 7 or 8, characterized in that, The first data generation and acquisition unit further includes: The first construction module is used to construct calibration trigger conditions corresponding to the real-time operating status of the vehicle; wherein, the calibration trigger conditions include: the vehicle speed is within the range of 10km / h to 120km / h; and the absolute value of the vehicle pitch angle is less than 0.2°; The third judgment module is used to determine whether the road surface is undulating based on the pitch angular velocity, pitch angle, and z-axis acceleration and displacement information output by the vehicle inertial navigation system. And / or, the data construction and determination generation unit further includes: The third generation module is used to generate and acquire sixth data corresponding to the real-time operating status of the vehicle and within the range of the radar, and to calculate and generate seventh data relative to the radar baseline based on the sixth data; wherein, the range of the radar is 10m to 150m; the sixth data is high-altitude static target data with a height of not less than 3m; the seventh data is altitude value data; the lowest point of the target point cloud cluster is taken as the reference point; The fourth determination module is used to determine the changing trend of the seventh data as the distance decreases in real time. If the seventh data increases as the distance decreases, it is determined that the radar is installed upward; if the seventh data decreases as the distance decreases, it is determined that the radar is installed downward.

10. A self-calibration processing platform for elevation angle of a 4D imaging millimeter-wave radar based on ground and high-altitude targets, characterized in that, The system includes a processor, a memory, and a control program for a 4D imaging millimeter-wave radar elevation angle self-calibration processing platform based on ground and high-altitude targets. The processor executes the control program, which is stored in the memory. This control program implements the 4D imaging millimeter-wave radar elevation angle self-calibration processing method for ground and high-altitude targets as described in any one of claims 1 to 6.

Images