Dynamic self-calibration processing method, system and platform for pitch angle of 4D imaging millimeter wave radar based on road guardrail

Abstract

The invention discloses a 4D imaging millimeter wave radar pitch angle dynamic self-calibration processing method, system and platform based on a road guardrail, and the method comprises the steps: employing a common road guardrail at the side of a vehicle as a calibration object, analyzing the height change rule of a guardrail point cloud detected by a radar in real time, employing an intelligent iteration compensation algorithm, and achieving the dynamic self-calibration of the pitch angle of a 4D imaging millimeter wave radar. On-line calibration of the pitch angle deviation of the 4D imaging millimeter wave radar can be automatically and accurately completed in the vehicle driving process, the problem that height measurement is not accurate due to the change of the radar installation angle is effectively solved, the recognition reliability of an advanced driving assistance system on air and ground targets is greatly improved, no extra calibration facility is needed, and the cost is reduced. The maintenance cost is obviously reduced; according to the method, a metal guardrail which is common on a road and easy to recognize by a radar is used as a calibration object, a pitch angle deviation compensation value is automatically calculated in an iterative compensation mode by analyzing a height change rule of a guardrail point cloud detected by the radar, complex calculation and special tools are not needed, and the method can be automatically completed in a vehicle driving process.

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 dynamic self-calibration of pitch angle of 4D imaging millimeter-wave radar based on road guardrails. Background Technology

[0002] With the widespread application of 77GHz automotive millimeter-wave radar in Advanced Driver Assistance Systems (ADAS), its detection accuracy directly determines the performance of functions such as Lane Keeping Assist (LKA), Forward Collision Warning (FCW), and Automatic Emergency Braking (AEB). The radar's ability to locate and measure the altitude of targets is crucial. For example, the system needs to accurately distinguish between aerial road signs and streetlights (passable targets) and ground vehicles and obstacles (non-passable targets) to avoid false triggering.

[0003] Traditionally, the range r and elevation angle are measured by radar. Calculate height value The radar is used to determine the target's height. However, there is an initial installation angle deviation when the radar is installed in the vehicle, and during the vehicle's lifespan, road bumps and vibrations may cause the installation angle to change slowly, resulting in deviations in pitch angle measurements and consequently inaccurate height calculations. Currently, the initial deviation is mainly compensated for by end-of-line (EOL) calibration, but EOL calibration requires precision instruments and specific locations. For vehicles already sold, recalling them to 4S stores for calibration due to angle deviations is costly.

[0004] In recent years, 4D imaging millimeter-wave radar technology has become increasingly widespread, offering more precise pitch angle measurement capabilities (accuracy up to ±0.1°), providing a hardware foundation for high-precision pitch angle self-calibration. While azimuth self-calibration technology is relatively mature, dynamic pitch angle self-calibration remains a relatively new challenge. Current technologies lack a simple, efficient pitch angle calibration solution that can be performed during routine vehicle operation without complex tools.

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

[0006] To overcome the shortcomings and difficulties of the existing technology, the purpose of this invention is to propose a dynamic self-calibration processing method, system and platform for the pitch angle of a 4D imaging millimeter-wave radar based on road guardrails. Therefore, in view of the above technical problems and defects, it is urgent to design and develop a dynamic self-calibration processing method, system and platform for the pitch angle of a 4D imaging millimeter-wave radar based on road guardrails.

[0007] The first objective of this invention is to provide a dynamic self-calibration processing method for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails; the second objective of this invention is to provide a dynamic self-calibration processing system for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails; and the third objective of this invention is to provide a dynamic self-calibration processing platform for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails.

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

[0009] The system generates and acquires first data corresponding to the real-time status of the vehicle, and generates corresponding second data based on the first data; wherein, the first data is data on the vehicle driving or stationary on a road without undulations, and the radar detects a strong reflective road guardrail that meets preset conditions; the second data is control data for the self-calibration program to start.

[0010] Generate and acquire third data corresponding to the first data, and calculate and generate fourth data corresponding to the third data. Create and determine a first change pattern corresponding to the fourth data. The third data consists of distance data and original pitch angle data of multiple reflection points on the guardrail. The fourth data consists of height value data of each reflection point. The first change pattern includes: a pattern where the height value gradually increases from near to far, a pattern where the height value gradually decreases from near to far, and a pattern where the height value remains basically unchanged.

[0011] Based on the first change pattern and combined with the first preset threshold, iterative compensation processing is performed to generate fifth data corresponding to the real-time status of the vehicle, and sixth data corresponding to the fifth data is calculated and generated; wherein, the first preset threshold is preset step size data; the fifth data is pitch angle data after compensation processing; and the sixth data is height value data of each reflection point after compensation processing.

[0012] Generate and acquire seventh data corresponding to the first change pattern, and generate and output corresponding eighth data based on the seventh data for subsequent radar angle measurement and correction processing; wherein, the seventh data is pitch angle deviation compensation value data.

[0013] Furthermore, the criteria for judging the strong reflective road guardrail data that meet the preset conditions include: the signal-to-noise ratio of the guardrail point cloud exceeds the first threshold, the estimated height of the guardrail exceeds the second threshold, and the distance between consecutive appearances of the guardrail exceeds the third threshold.

[0014] Wherein, the first threshold is the signal-to-noise ratio threshold; the second threshold is the guardrail height threshold; and the third threshold is the guardrail continuous length threshold.

[0015] Furthermore, the determination of a road without undulations is based on the attitude angular velocity information output by the vehicle's inertial navigation unit; wherein, when the angular velocity is lower than a set threshold, the current road segment is determined to be a road segment without undulations.

[0016] Furthermore, the step of iteratively compensating and generating fifth data corresponding to the real-time state of the vehicle based on the first change pattern and combined with a first preset threshold, and calculating and generating sixth data corresponding to the fifth data, further includes:

[0017] If the initial judgment of the height change conforms to the rule that the height value gradually decreases from near to far in the first change rule, then a positive step compensation process is performed until the first change rule changes to the rule that the height value gradually increases from near to far.

[0018] If the initial judgment of the height change conforms to the first change pattern where the height value gradually increases from near to far, then negative step compensation processing is performed until the change pattern changes to a pattern where the height value gradually decreases from near to far.

[0019] Furthermore, the preset step size of the step compensation is 0.1°; wherein the preset cumulative compensation angle range is ±3°.

[0020] Furthermore, the iterative compensation process employs a variable step size strategy; initially, a larger first step size is used for compensation, and when the change pattern reverses, a smaller second step size is used for reverse or fine compensation.

[0021] Furthermore, after generating and acquiring the seventh data corresponding to the first change pattern, and generating and outputting the corresponding eighth data based on the seventh data for subsequent radar angle measurement and correction processing, the method further includes:

[0022] Using the determined cumulative compensation value data, calculate and generate guardrail height value data at multiple preset location points at far-field distances;

[0023] Determine whether the height difference of the corresponding far-field position points is within the allowable tolerance range. If so, the calibration is confirmed to be successful; otherwise, fine-tune the cumulative compensation value and verify it again.

[0024] The second objective of this invention is achieved as follows: the system is used to implement the dynamic self-calibration processing method for the elevation angle of the 4D imaging millimeter-wave radar based on road guardrails, the system comprising:

[0025] The first data generation and acquisition unit is used to generate and acquire first data corresponding to the real-time status of the vehicle, and generate corresponding second data based on the first data; wherein, the first data is data of the vehicle driving or stationary on a road without undulations, and the radar detects a strong reflective road guardrail that meets preset conditions; the second data is self-calibration program start control data;

[0026] The second data generation and acquisition unit is used to generate and acquire third data corresponding to the first data, calculate and generate fourth data corresponding to the third data, and create and determine a first change pattern corresponding to the fourth data; wherein, the third data is distance data and original pitch angle data of multiple reflection points on the guardrail; the fourth data is height value data of each reflection point; the first change pattern includes: the height value gradually increases from near to far, the height value gradually decreases from near to far, and the height value remains basically unchanged;

[0027] The data iterative compensation processing unit is used to iteratively compensate and generate fifth data corresponding to the real-time state of the vehicle based on the first change law and combined with the first preset threshold, and to calculate and generate sixth data corresponding to the fifth data; wherein, the first preset threshold is preset step size data; the fifth data is pitch angle data after compensation processing; and the sixth data is height value data of each reflection point after compensation processing.

[0028] The data generation and output unit is used to generate and acquire the seventh data corresponding to the first change law, and generate and output the corresponding eighth data based on the seventh data for subsequent radar angle measurement and correction processing; wherein, the seventh data is pitch angle deviation compensation value data.

[0029] Furthermore, the criteria for judging the strong reflective road guardrail data that meet the preset conditions include: the signal-to-noise ratio of the guardrail point cloud exceeds the first threshold, the estimated height of the guardrail exceeds the second threshold, and the distance between consecutive appearances of the guardrail exceeds the third threshold.

[0030] Wherein, the first threshold is a signal-to-noise ratio threshold; the second threshold is a guardrail height threshold; and the third threshold is a guardrail continuous length threshold.

[0031] The determination of a road without undulations is based on the attitude angular velocity information output by the vehicle's inertial navigation unit; when the angular velocity is lower than a set threshold, the current road segment is determined to be a road segment without undulations.

[0032] The data iteration compensation processing unit further includes:

[0033] The first determination module is used to perform positive step compensation processing if the initial determination of the height change conforms to the rule that the height value gradually decreases from near to far in the first change rule, until the first change rule changes to the rule that the height value gradually increases from near to far.

[0034] The second determination module is used to perform negative step compensation processing if the initial determination of the height change conforms to the first change law where the height value gradually increases from near to far, until the change law changes to the law where the height value gradually decreases from near to far.

[0035] The preset step size for the step compensation is 0.1°; wherein the preset cumulative compensation angle range is ±3°.

[0036] The iterative compensation process employs a variable step size strategy; initially, a larger step size is used for compensation, and when the change pattern reverses, a smaller step size is used for reverse or fine compensation.

[0037] The system also includes:

[0038] The first generation module is used to calculate and generate guardrail height data at multiple preset location points at far-field distances using determined cumulative compensation value data;

[0039] The third judgment module is used to determine whether the height difference of the corresponding far-field position point is within the allowable tolerance range. If so, the calibration is confirmed to be successful; otherwise, the cumulative compensation value is fine-tuned and then verified again.

[0040] 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 pitch angle dynamic self-calibration processing platform based on road guardrails; wherein the processor executes the control program for the 4D imaging millimeter-wave radar pitch angle dynamic self-calibration processing platform based on road guardrails, the control program is stored in the memory, and the control program implements the 4D imaging millimeter-wave radar pitch angle dynamic self-calibration processing method based on road guardrails.

[0041] This invention generates and acquires first data corresponding to the real-time status of a vehicle through a method, and generates corresponding second data based on the first data; wherein, the first data is data of a strong-reflection road guardrail detected by radar when the vehicle is driving or stationary on a flat road, and meets preset conditions; the second data is self-calibration program start-up control data; generates and acquires third data corresponding to the first data, calculates and generates fourth data corresponding to the third data, and creates and determines a first change law corresponding to the fourth data; wherein, the third data is distance data and original pitch angle data of multiple reflection points on the guardrail; the fourth data is height value data of each reflection point; the first change law includes: a law of gradually increasing height from near to far, a law of gradually decreasing height from near to far, and a law of basically unchanged height; based on the first change law, combined with a first preset threshold, The system iteratively compensates and generates fifth data corresponding to the real-time vehicle status, and calculates and generates sixth data corresponding to the fifth data; wherein, the first preset threshold is preset step size data; the fifth data is pitch angle data after compensation processing; the sixth data is height value data of each reflection point after compensation processing; seventh data corresponding to the first change law is generated and obtained, and eighth data corresponding to the seventh data and used for subsequent radar angle measurement and correction processing is generated and output based on the seventh data; wherein, the seventh data is pitch angle deviation compensation value data, and a system and platform corresponding to the method are also provided. By using common and easily identifiable metal guardrails on the road as calibration objects, and by analyzing the height change law of the radar detection guardrail point cloud, the pitch angle deviation compensation value is automatically calculated by iterative compensation, without the need for complex calculations and special tools, and can be completed automatically during vehicle operation.

[0042] In other words, the present invention utilizes common road guardrails on the side of vehicles as calibration objects, analyzes the height change patterns of guardrail point clouds detected by radar in real time, and adopts an intelligent iterative compensation algorithm to automatically and accurately complete the online calibration of the pitch angle deviation of 4D imaging millimeter-wave radar during vehicle operation. This effectively solves the problem of inaccurate height measurement caused by changes in radar installation angle, significantly improves the reliability of advanced driver assistance systems in identifying air and ground targets, and requires no additional calibration facilities, thus significantly reducing maintenance costs. Attached Figure Description

[0043] 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.

[0044] Figure 1Schematic diagram of existing vehicle-mounted millimeter-wave radar for measuring elevation angle / height;

[0045] Figure 2 (a) is a schematic diagram of dynamic pitch calibration of a 4D imaging millimeter-wave radar pitch angle dynamic self-calibration processing method based on road guardrail according to the present invention.

[0046] (b) is a schematic diagram of the elevation of the radar to the highest point of the guardrail in the plane R'-P1-O of the present invention, which is a dynamic self-calibration processing method for the elevation angle of a 4D imaging millimeter-wave radar based on a road guardrail.

[0047] Figure 3 This is a schematic diagram of radar height / elevation measurement on the plane where the guardrail is located, illustrating the dynamic self-calibration processing method for elevation angle of a 4D imaging millimeter-wave radar based on road guardrails according to the present invention.

[0048] Figure 4 This is a schematic diagram of the radar measuring the height / elevation angle of the highest point of the guardrail in the plane where the guardrail is located, when the installation elevation plane of the radar is tilted upward by an angle θ. This is a dynamic self-calibration processing method for the elevation angle of a 4D imaging millimeter-wave radar based on a road guardrail according to the present invention.

[0049] Figure 5 This is a schematic diagram of the radar measuring the height / elevation angle of the highest point of the guardrail in the plane where the guardrail is located, when the radar's installation elevation plane is tilted downward by an angle θ. This is a dynamic self-calibration processing method for the elevation angle of a 4D imaging millimeter-wave radar based on a road guardrail according to the present invention.

[0050] Figure 6 (a) is a schematic diagram of the radar measuring the height / elevation angle of the highest point of the guardrail in the plane where the guardrail is located when the installation elevation plane of the radar is tilted upward by an angle θ according to the dynamic self-calibration processing method of 4D imaging millimeter wave radar based on road guardrail of the present invention.

[0051] (b) is a schematic diagram of the radar measuring the height / elevation angle of the lowest point of the guardrail in the plane where the guardrail is located when the installation elevation plane of the radar is tilted upward by an angle θ according to the dynamic self-calibration processing method of 4D imaging millimeter wave radar based on road guardrail of the present invention.

[0052] Figure 7 This is a schematic diagram of the process steps of a dynamic self-calibration method for elevation angle of a 4D imaging millimeter-wave radar based on road guardrails according to the present invention.

[0053] Figure 8 This is a schematic diagram of the architecture of a 4D imaging millimeter-wave radar pitch angle dynamic self-calibration processing system based on road guardrails according to the present invention.

[0054] Figure 9This is a schematic diagram of the dynamic self-calibration processing platform architecture for a 4D imaging millimeter-wave radar elevation angle based on road guardrails, according to the present invention. Detailed Implementation

[0055] 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.

[0056] 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.

[0057] 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.

[0058] 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.

[0059] Preferably, the 4D imaging millimeter-wave radar elevation angle dynamic self-calibration processing method based on road guardrails 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.

[0060] 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.

[0061] This invention provides a method, system, and platform for dynamic self-calibration of elevation angle of 4D imaging millimeter-wave radar based on road guardrails.

[0062] like Figure 7 The diagram shown is a flowchart of a dynamic self-calibration processing method for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails, provided in an embodiment of the present invention.

[0063] In this embodiment, the dynamic self-calibration processing method for pitch angle of 4D imaging millimeter-wave radar based on road guardrails 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.

[0064] The dynamic self-calibration processing method for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails 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, a wide area network (WAN), a metropolitan area network (MAN), or a local area network (LAN). The dynamic self-calibration processing method for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails in this embodiment of the invention can be executed by the server, by the terminal, or by both the server and the terminal.

[0065] For example, for a terminal requiring dynamic self-calibration of the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails, the method of this invention can be directly integrated into the terminal to provide the dynamic self-calibration function, or a client application 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 as a Software Development Kit (SDK), providing an interface for the dynamic self-calibration function of the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails. The terminal or other devices can then implement this function through the provided interface. The invention will be further described below with reference to the accompanying drawings.

[0066] like Figure 7 As shown, this invention provides a dynamic self-calibration method for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails.

[0067] S1. Generate and acquire first data corresponding to the real-time status of the vehicle, and generate corresponding second data based on the first data; wherein, the first data is data of the vehicle driving or stationary on a road without undulations, and the radar detects a strong reflective road guardrail that meets preset conditions; the second data is self-calibration program start control data;

[0068] S2. Generate and acquire third data corresponding to the first data, and calculate and generate fourth data corresponding to the third data, and create and determine a first change pattern corresponding to the fourth data; wherein, the third data is the distance data and original pitch angle data of multiple reflection points on the guardrail; the fourth data is the height value data of each reflection point; the first change pattern includes: the height value gradually increases from near to far, the height value gradually decreases from near to far, and the height value remains basically unchanged;

[0069] S3. Based on the first change law and combined with the first preset threshold, iteratively compensate and generate the fifth data corresponding to the real-time state of the vehicle, and calculate and generate the sixth data corresponding to the fifth data; wherein, the first preset threshold is the preset step size data; the fifth data is the pitch angle data after compensation processing; the sixth data is the height value data of each reflection point after compensation processing;

[0070] S4. Generate and acquire the seventh data corresponding to the first change law, and generate and output the corresponding eighth data based on the seventh data for subsequent radar angle measurement and correction processing; wherein, the seventh data is pitch angle deviation compensation value data.

[0071] The criteria for determining the strong reflective road guardrail data that meets the preset conditions include: the signal-to-noise ratio of the guardrail point cloud exceeds the first threshold, the estimated height of the guardrail exceeds the second threshold, and the distance between consecutive appearances of the guardrail exceeds the third threshold.

[0072] Wherein, the first threshold is the signal-to-noise ratio threshold; the second threshold is the guardrail height threshold; and the third threshold is the guardrail continuous length threshold.

[0073] The determination of a road without undulations is based on the attitude angular velocity information output by the vehicle's inertial navigation unit; when the angular velocity is lower than a set threshold, the current road segment is determined to be a road segment without undulations.

[0074] The process of iteratively compensating and generating fifth data corresponding to the real-time state of the vehicle based on the first change pattern and combined with a first preset threshold, and calculating and generating sixth data corresponding to the fifth data, further includes:

[0075] S31. If the initial judgment of the height change conforms to the rule that the height value gradually decreases from near to far in the first change rule, then perform positive step compensation processing until the first change rule changes to the rule that the height value gradually increases from near to far.

[0076] S32. If the initial judgment of the height change conforms to the first change rule of the height value gradually increasing from near to far, then negative step compensation processing is performed until the change rule changes to the rule of the height value gradually decreasing from near to far.

[0077] The preset step size for the step compensation is 0.1°; wherein the preset cumulative compensation angle range is ±3°.

[0078] The iterative compensation process employs a variable step size strategy; initially, a larger step size is used for compensation, and when the change pattern reverses, a smaller step size is used for reverse or fine compensation.

[0079] After generating and acquiring the seventh data corresponding to the first change pattern, and generating and outputting the corresponding eighth data based on the seventh data for subsequent radar angle measurement and correction processing, the method further includes:

[0080] S50. Using the determined cumulative compensation value data, calculate and generate guardrail height value data for multiple preset location points at far-field distances;

[0081] S60. Determine whether the height difference of the corresponding far-field position points is within the allowable tolerance range. If so, confirm that the calibration is successful; otherwise, fine-tune the cumulative compensation value and verify it again.

[0082] Specifically, in an embodiment of the present invention, a method for dynamic self-calibration of the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails is provided, comprising the following steps:

[0083] S01: Trigger Condition Judgment: When the vehicle is driving or stationary on a flat road and the radar detects a strong reflective road guardrail that meets preset conditions, the self-calibration program is initiated; S02: Data Acquisition and Processing: Within a preset near-field distance range, the distance (r) and original pitch angle (α) data of multiple reflection points on the guardrail are acquired, and the height value of each point is calculated (H = r * sinα); S03: Change Pattern Judgment: The change pattern of the maximum height of the guardrail point cloud within each distance unit from near to far is statistically analyzed and judged. The change pattern includes: the height value gradually increases from near to far (pattern A), the height value... The elevation value gradually decreases from near to far (rule B) or remains basically unchanged; S04: Iterative compensation: Based on the change rule, the original elevation angle (α) is iteratively compensated with a preset step size (Δ) to obtain the compensated elevation angle (α'), and the elevation value (H'=r*sinα') is recalculated until the change rule of the maximum elevation value is reversed or the cumulative compensation angle exceeds the preset range; S05: Compensation value confirmation and output: When the change rule is reversed, the current cumulative compensation value (Δα) is determined as the elevation angle deviation compensation value and output for subsequent radar angle measurement correction.

[0084] The criteria for determining a strong reflective road guardrail that meets preset conditions in step S01 include: the signal-to-noise ratio of the guardrail point cloud exceeding a first threshold, the estimated height of the guardrail exceeding a second threshold, and the distance between consecutive appearances of the guardrail exceeding a third threshold. The criteria for determining a road without undulations in step S01 are derived from the attitude angular velocity information output by the vehicle's inertial navigation unit (IMU). When the angular velocity is lower than a set threshold, the current road segment is determined to be a road without undulations. The specific process of iterative compensation in step S04 is as follows: if the initial judgment of height change conforms to rule B, then positive step compensation (+Δ) is performed until the change rule changes to rule A; if the initial judgment of height change conforms to rule A, then negative step compensation (-Δ) is performed until the change rule changes to rule B. The preset step size (Δ) is 0.1°, and the preset cumulative compensation angle range is ±3°. In step S04, a variable step size strategy is used for iterative compensation: initially, a larger step size is used for compensation, and when the change pattern reverses, a smaller step size is used for reverse or fine compensation to accelerate the convergence process.

[0085] Following step S05, a verification step is also included: S06: Using the determined cumulative compensation value (Δα), calculate the guardrail height values ​​at multiple preset location points at the far-field distance; S07: Determine whether the height difference between these far-field location points is within the allowable tolerance range (Δh). If yes, the calibration is confirmed to be successful; if not, the cumulative compensation value (Δα) is fine-tuned and verified again. When calculating the height value (H = r * sinα), a lookup table method is used to obtain the sine function value. The lookup table pre-stores sine values ​​within the expected compensation angle range, and the angle interval matches the preset step size (Δ). The guardrail point cloud is either the highest or lowest point cloud of the guardrail, and the height variation law corresponds to the height variation law of the highest point or the height variation law of the lowest point.

[0086] In other words, the 4D radar pitch self-calibration of this invention, simply put, involves the radar measuring the pitch angle / height of the side guardrail point cloud when the vehicle is on a level road. The guardrail must be a highly reflective metal guardrail, of uniform height, and not too low; the guardrail height is typically between 90cm and 120cm. This allows the vehicle to estimate whether the guardrail height meets the requirements by detecting the highest point of the guardrail at a close distance (approximately 2m to 4m) while driving in the lane closest to the guardrail. Based on the guardrail height information measured within a set distance range, the radar performs a calibration process, continuously adjusting the compensation value towards a set target, ultimately completing the pitch angle calibration.

[0087] The vehicle is traveling or stationary on a flat road. Assume the 4D radar R is installed directly in front of the vehicle, approximately 30cm above the ground. The radar measures the point cloud of the guardrail, and the highest points measured are P1, P2, ... these reflection points. Typically, the maximum height of metal guardrails on roads is between 90cm and 120cm, and the maximum height of the guardrail point cloud normally measured by the radar also falls within this range.

[0088] The point cloud of guardrails, especially metal guardrails, appears as a coarse and dense cluster of highly reflective point clouds in the 4D imaging radar detection points, as shown in the newly added schematic diagram on the left.

[0089] The guardrail recognition module in the radar data processing program will fit the guardrail curve based on the point cloud distribution on both sides of the vehicle. Points within a certain distance range on both sides of the guardrail curve (e.g., within 0.5m on both sides) will be marked with guardrail markers, and thus be recognized by the radar as guardrail points.

[0090] When calibrating, select guardrail point clouds within a distance with a high signal-to-noise ratio. For example, if the radar can detect up to 300m, guardrail points within 200m can be selected.

[0091] Projecting the radar position R onto the plane O-P1-P4 where the guardrail is located, we can obtain the following: Figure 2 (b) or Figure 3 The view shown. Figure 3 The upper part is the view of the guardrail as "seen" by the radar, that is, a schematic diagram of elevation angle / height measurement obtained with the radar antenna surface normal as the horizontal baseline.

[0092] Figure 3 The dashed line represents the normal direction of the radar antenna surface, which is normally parallel to the ground. It is also the radar's default horizontal line and the baseline for measuring elevation angles; the elevation angles measured by the radar are relative to this dashed line. If the radar antenna surface is no longer perpendicular to the ground but tilted upwards at an angle θ, the radar's viewing angle of the guardrail changes. Figure 4 As shown. Figure 4 The upper part is the view of the guardrail as "seen" by the radar, that is, the elevation angle / height measurement view obtained with the radar antenna surface normal as the horizontal baseline: the maximum height of the guardrail measured by the radar gradually decreases from near to far.

[0093] When the radar elevation plane tilts upwards by an angle θ, the radar's default "horizontal line" also tilts by an angle θ. According to... Figure 4 According to the geometric relationship, the angle between the straight line fitted at the highest point of the guardrail and the dashed line is θ.

[0094] Therefore, the most direct approach is to first measure the distance and elevation angle information of the point cloud at the highest point of each guardrail, then project the radar center onto the plane containing the guardrail, and simultaneously perform a similar projection transformation on the distance and elevation angle information of the point cloud at the highest point of the guardrail. Finally, on the plane containing the guardrail, calculate the angle θ between the straight line fitted from the highest point of the guardrail and the normal direction of the radar antenna surface. This is the elevation deviation angle that needs to be compensated.

[0095] This scheme does not intend to use the above-mentioned method of directly calculating the pitch calibration angle through geometric projection. Instead, it bypasses the calculation of the deviation angle based on a simple rule and obtains the calibration result more directly.

[0096] Suppose the radar antenna is tilted downwards at an angle θ. At this point, the radar's viewing angle towards the guardrail becomes... Figure 5 As shown. Figure 5 The upper part is the view of the guardrail as "seen" by the radar, which is a schematic diagram of elevation angle measurement / altitude measurement obtained with the radar elevation plane normal as the horizontal reference line: the maximum height of the guardrail measured by the radar gradually increases from near to far.

[0097] The variation in guardrail height measured by the radar antenna at different tilt angles shows that the maximum height of the guardrail measured by the radar is the same only when the radar antenna is perpendicular to the ground. Let A be the fact that "the maximum height of the guardrail measured by the radar gradually increases from near to far," and let B be the fact that "the maximum height of the guardrail measured by the radar gradually decreases from near to far."

[0098] The pitch self-calibration algorithm flow is designed as follows:

[0099] 1. When the vehicle's radar detects a long stretch (e.g., more than 100m) of highly reflective guardrail on a flat road surface, the degree of road undulation can be determined based on the angular velocity information from the vehicle's attitude offset data, which is output in real-time by the vehicle's inertial navigation (IMU) chip to the domain controller. Highly reflective guardrails can be identified by statistically analyzing the signal-to-noise ratio (SNR) of guardrail points within a certain distance range. For example, if over 90% of guardrail points at a specified distance of 50m-55m have a SNR greater than 30dB, then the guardrail is considered a highly reflective guardrail. The vehicle is then driven or stopped in the lane closest to the guardrail. The guardrail height (requirement >90cm) is estimated by detecting the highest point of the guardrail at close range (approximately 2m to 4m). If the height meets the minimum calibration requirements, the radar pitch self-calibration procedure is initiated.

[0100] 2. The radar collects elevation angle information of guardrail points within each range cell from near to far: The distance *r* and elevation angle *α* of the guardrail point cloud measured by the radar are collected, with *r* ranging from 2m to 50m. The height of the guardrail point is calculated using the formula H = *r*sinα. Simultaneously, the highest elevation values ​​of the guardrail point cloud near *r* equal to 50m, 75m, and 100m are collected and stored.

[0101] 3. Statistically count and store the highest point values ​​of the guardrail point cloud within the range r from 2m to 50m. When the highest point value of the guardrail meets the criteria B, compensate the radar elevation angle by +0.1°—that is, add 0.1° to the elevation angle of all the highest point points of the guardrail within the statistical range r from 2m to 50m. Then recalculate the statistical value of the highest point of the guardrail using the formula H = r * sin(α'), where α' is the compensated elevation angle. If the highest point value of the guardrail still meets the criteria B, continue to compensate the elevation angle by +0.1°. The program ends when the highest point value of the guardrail changes to meet the criteria A or the cumulative compensated angle value is >3°. If the sum of the cumulative compensated angles is ≤3°, the highest point value of the guardrail changes from meeting the criteria B to meeting the criteria A, then proceed to step 4. If the cumulative compensated angle value is >3° and has not yet changed to meet the criteria A, it indicates that the radar installation deviation angle exceeds the calibrable range, and 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. (Generally, car manufacturers stipulate that radars with an installation angle deviation exceeding ±3°, and at most ±4°, must be returned to a 4S shop for repair. The radar must be disassembled and reinstalled before EOL calibration or self-calibration is performed. This is a mandatory requirement for all OEMs. Some OEMs have more lenient requirements, allowing a maximum deviation of 4°. Some have stricter requirements, allowing only 3°.)

[0102] 4. There are many reasons why OEMs specify this, some of which are speculated as follows: For automotive radar, the pitch field of view (FOV) is typically ±15°. A deviation of 3° already reduces the usable FOV range to 90% of the nominal value. This is equivalent to a height deviation of r*sin3° or 2.6 meters for targets 20 meters away. This is enough to misjudge traffic lights at high altitudes as ground obstacles, thus triggering AEB braking.

[0103] The highest point value of the guardrail point cloud within the range r from 2m to 50m is statistically analyzed. When the highest point value of the guardrail meets the condition A, the radar elevation angle is compensated by -0.1°—that is, the elevation angle of all guardrail highest point point clouds within the statistical range r from 2m to 50m is subtracted by 0.1°. Then, the value of the highest point of the guardrail is recalculated using the formula H = r * sin(α'), where α' is the compensated elevation angle. If the highest point value of the guardrail still meets the condition A, the elevation angle is compensated by -0.1°. The program ends when the highest point value of the guardrail changes to meet the condition B or the cumulative compensated angle value is < -3°. If the sum of the cumulative compensated angles is ≥ -3°, the highest point value of the guardrail changes from meeting the condition A to meeting the condition B, then proceed to step 4. If the cumulative compensated angle value is < -3° and has not yet changed to meet the condition B, it indicates that the radar installation deviation angle exceeds the calibrable range, and 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.

[0104] For the cumulative compensation value Δα obtained in step 3, the highest values ​​of the guardrail points near the radar at distances of 50m, 75m, and 100m, stored in step 2, are recalculated using the formula H = r * sin(α + Δα). A tolerance Δh is set (e.g., 20cm, determined experimentally based on actual radar system parameters). [This is because the radar has limitations due to angular measurement error Δθ caused by angular measurement accuracy and ranging error Δr caused by ranging accuracy, for example, angular measurement accuracy fluctuates between ±0.1° and ±0.2°.] The measured height h + Δh = (r + Δr) * sin(θ + Δθ) = sinθcosΔθ + cosθsinΔθ. Because Δθ is very small, sinθcosΔθ + cosθsinΔθ is approximately sinθ + Δθ * cosθ.

[0105] So h+Δh=(r+Δr)*(sinθ+Δθ*cosθ)

[0106] Given h = r * sinθ, then Δh = r * Δθ * cosθ + Δr * sinθ + Δr * Δθ * cosθ. Since θ is also relatively small at long distances, ignoring higher-order minor quantities, Δh ≈ r * Δθ + Δr * θ. Therefore, the larger r is, the larger Δh is. Thus, for guardrails with r of 75m and 100m, it is necessary to set a tolerance Δh. Determine whether the condition "the height difference of the highest point of the guardrail measured by radar does not exceed Δh" is met at 50m, 75m, and 100m. Compare the height difference of the highest point of the guardrail near 50m, 75m, and 100m when the cumulative compensation value is Δα with that when (Δα - 0.1°) is used: if the height difference of the highest point of the guardrail at 50m, 75m, and 100m is smaller after compensation of Δα than after compensation of (Δα - 0.1°), then the cumulative compensation value is taken as Δα; otherwise, (Δα - 0.1°) is used. The calibration ends upon successful calibration.

[0107] Additional explanation of the above process: 0.1° was chosen as the compensation step and minimum interval because the angular measurement accuracy of the 4D radar is approximately ±0.1°. Smaller steps would exceed the radar's accuracy range and would therefore be meaningless.

[0108] The method described above, which searches for angle compensation values ​​at 0.1° intervals, can be improved to a dynamic step-size adjustment strategy. Initially, a larger step size is used (0.1° to 1°, minimum interval 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 from A to B or from B to A 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, the compensation is increased by 1°. This iterative approach accelerates convergence. The Python code describing the dynamic step-size adjustment is as follows:

[0109]

[0110]

[0111]

[0112] The calculation of the sine function involved in calculating the guardrail height can be improved by using a lookup table to speed up the calculation. Sine function values ​​within the angle range used for elevation angle calibration can be pre-stored, such as sin(-5°), sin(-4.9°), ... up to sin(5°), in 0.1° intervals, because the radar angle measurement accuracy is 0.1°. The calculation of H = r * sinα can then be directly obtained from the table.

[0113] Finally, since the highest point of the guardrail can be measured to determine the pitch angle, it is naturally also possible to measure the lowest point (ground) of the guardrail to determine the pitch angle. (As above) Figure 6As shown in (b), if the elevation angle is tilted upwards by an angle θ, the absolute value of the lowest point of the guardrail measured by the radar becomes larger from near to far. Conversely, it becomes smaller. Similar to calibrating the elevation angle by measuring the highest point of the guardrail, this rule can be used to design a similar elevation calibration algorithm.

[0114] To achieve the above objectives, the present invention also provides a dynamic self-calibration processing system for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails, such as... Figure 8 As shown, the system is used to implement the dynamic self-calibration processing method for the elevation angle of the 4D imaging millimeter-wave radar based on road guardrails. The system includes:

[0115] The first data generation and acquisition unit is used to generate and acquire first data corresponding to the real-time status of the vehicle, and generate corresponding second data based on the first data; wherein, the first data is data of the vehicle driving or stationary on a road without undulations, and the radar detects a strong reflective road guardrail that meets preset conditions; the second data is self-calibration program start control data;

[0116] The second data generation and acquisition unit is used to generate and acquire third data corresponding to the first data, calculate and generate fourth data corresponding to the third data, and create and determine a first change pattern corresponding to the fourth data; wherein, the third data is distance data and original pitch angle data of multiple reflection points on the guardrail; the fourth data is height value data of each reflection point; the first change pattern includes: the height value gradually increases from near to far, the height value gradually decreases from near to far, and the height value remains basically unchanged;

[0117] The data iterative compensation processing unit is used to iteratively compensate and generate fifth data corresponding to the real-time state of the vehicle based on the first change law and combined with the first preset threshold, and to calculate and generate sixth data corresponding to the fifth data; wherein, the first preset threshold is preset step size data; the fifth data is pitch angle data after compensation processing; and the sixth data is height value data of each reflection point after compensation processing.

[0118] The data generation and output unit is used to generate and acquire the seventh data corresponding to the first change law, and generate and output the corresponding eighth data based on the seventh data for subsequent radar angle measurement and correction processing; wherein, the seventh data is pitch angle deviation compensation value data.

[0119] The criteria for determining the strong reflective road guardrail data that meets the preset conditions include: the signal-to-noise ratio of the guardrail point cloud exceeds the first threshold, the estimated height of the guardrail exceeds the second threshold, and the distance between consecutive appearances of the guardrail exceeds the third threshold.

[0120] Wherein, the first threshold is a signal-to-noise ratio threshold; the second threshold is a guardrail height threshold; and the third threshold is a guardrail continuous length threshold.

[0121] The determination of a road without undulations is based on the attitude angular velocity information output by the vehicle's inertial navigation unit; when the angular velocity is lower than a set threshold, the current road segment is determined to be a road segment without undulations.

[0122] The data iteration compensation processing unit further includes:

[0123] The first determination module is used to perform positive step compensation processing if the initial determination of the height change conforms to the rule that the height value gradually decreases from near to far in the first change rule, until the first change rule changes to the rule that the height value gradually increases from near to far.

[0124] The second determination module is used to perform negative step compensation processing if the initial determination of the height change conforms to the first change law where the height value gradually increases from near to far, until the change law changes to the law where the height value gradually decreases from near to far.

[0125] The preset step size for the step compensation is 0.1°; wherein the preset cumulative compensation angle range is ±3°.

[0126] The iterative compensation process employs a variable step size strategy; initially, a larger step size is used for compensation, and when the change pattern reverses, a smaller step size is used for reverse or fine compensation.

[0127] The system also includes:

[0128] The first generation module is used to calculate and generate guardrail height data at multiple preset location points at far-field distances using determined cumulative compensation value data;

[0129] The third judgment module is used to determine whether the height difference of the corresponding far-field position point is within the allowable tolerance range. If so, the calibration is confirmed to be successful; otherwise, the cumulative compensation value is fine-tuned and then verified again.

[0130] In the system embodiment of the present invention, the specific details of the method steps involved in the dynamic self-calibration of the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails 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.

[0131] To achieve the above objectives, the present invention also provides a dynamic self-calibration processing platform for the elevation angle of a 4D imaging millimeter-wave radar based on road guardrails, such as... Figure 9As shown, the system includes a processor, a memory, and a control program for a 4D imaging millimeter-wave radar elevation angle dynamic self-calibration processing platform based on road guardrails. 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 dynamic self-calibration processing method based on road guardrails. For example:

[0132] S1. Generate and acquire first data corresponding to the real-time status of the vehicle, and generate corresponding second data based on the first data; wherein, the first data is data of the vehicle driving or stationary on a road without undulations, and the radar detects a strong reflective road guardrail that meets preset conditions; the second data is self-calibration program start control data;

[0133] S2. Generate and acquire third data corresponding to the first data, and calculate and generate fourth data corresponding to the third data, and create and determine a first change pattern corresponding to the fourth data; wherein, the third data is the distance data and original pitch angle data of multiple reflection points on the guardrail; the fourth data is the height value data of each reflection point; the first change pattern includes: the height value gradually increases from near to far, the height value gradually decreases from near to far, and the height value remains basically unchanged;

[0134] S3. Based on the first change law and combined with the first preset threshold, iteratively compensate and generate the fifth data corresponding to the real-time state of the vehicle, and calculate and generate the sixth data corresponding to the fifth data; wherein, the first preset threshold is the preset step size data; the fifth data is the pitch angle data after compensation processing; the sixth data is the height value data of each reflection point after compensation processing;

[0135] S4. Generate and acquire the seventh data corresponding to the first change law, and generate and output the corresponding eighth data based on the seventh data for subsequent radar angle measurement and correction processing; wherein, the seventh data is pitch angle deviation compensation value data.

[0136] The specific details of the steps have been explained above and will not be repeated here.

[0137] In this embodiment of the invention, the built-in processor of the 4D imaging millimeter-wave radar pitch angle dynamic self-calibration processing platform based on road guardrails 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 pitch angle dynamic self-calibration based on road guardrails.

[0138] The memory is used to store program code and various data. It is installed in the 4D imaging millimeter-wave radar pitch angle dynamic self-calibration processing platform based on road guardrails, and can complete the storage and retrieval of programs or data at high speed and automatically during operation.

[0139] 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.

[0140] This invention generates and acquires first data corresponding to the real-time status of a vehicle through a method, and generates corresponding second data based on the first data; wherein, the first data is data of a strong-reflection road guardrail detected by radar when the vehicle is driving or stationary on a flat road, and meets preset conditions; the second data is self-calibration program start-up control data; generates and acquires third data corresponding to the first data, calculates and generates fourth data corresponding to the third data, and creates and determines a first change law corresponding to the fourth data; wherein, the third data is distance data and original pitch angle data of multiple reflection points on the guardrail; the fourth data is height value data of each reflection point; the first change law includes: a law of gradually increasing height from near to far, a law of gradually decreasing height from near to far, and a law of basically unchanged height; based on the first change law, combined with a first preset threshold, The system iteratively compensates and generates fifth data corresponding to the real-time vehicle status, and calculates and generates sixth data corresponding to the fifth data; wherein, the first preset threshold is preset step size data; the fifth data is pitch angle data after compensation processing; the sixth data is height value data of each reflection point after compensation processing; seventh data corresponding to the first change law is generated and obtained, and eighth data corresponding to the seventh data and used for subsequent radar angle measurement and correction processing is generated and output based on the seventh data; wherein, the seventh data is pitch angle deviation compensation value data, and a system and platform corresponding to the method are also provided. By using common and easily identifiable metal guardrails on the road as calibration objects, and by analyzing the height change law of the radar detection guardrail point cloud, the pitch angle deviation compensation value is automatically calculated by iterative compensation, without the need for complex calculations and special tools, and can be completed automatically during vehicle operation.

[0141] In other words, the present invention utilizes common road guardrails on the side of vehicles as calibration objects, analyzes the height change patterns of guardrail point clouds detected by radar in real time, and adopts an intelligent iterative compensation algorithm to automatically and accurately complete the online calibration of the pitch angle deviation of 4D imaging millimeter-wave radar during vehicle operation. This effectively solves the problem of inaccurate height measurement caused by changes in radar installation angle, significantly improves the reliability of advanced driver assistance systems in identifying air and ground targets, and requires no additional calibration facilities, thus significantly reducing maintenance costs.

[0142] In other words, the advantages of this invention are: simplicity and ease of implementation, eliminating complex geometric derivations and calculations, and eliminating the need to directly calculate the calibration compensation angle; it has a high-speed convergence method and table lookup to accelerate calculation; it has low requirements for the calibration scene and strong adaptability. As long as the road surface is flat and the guardrail meets certain height restrictions and is of uniform height, calibration can be performed on curves, in driving or stationary states. Specifically, by utilizing the easily identifiable radar scene of road guardrails, a simple and easy-to-implement scheme is proposed for calibrating the 4D radar elevation angle by statistically analyzing the changes in guardrail height from near to far as measured by radar, and gradually adjusting the compensation angle accordingly.

[0143] 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 road fence-based 4D imaging millimeter wave radar pitch angle dynamic self-calibration processing method, characterized in that, The method comprises the steps of: generating and obtaining first data corresponding to the real-time state of the vehicle, and generating corresponding second data based on the first data; wherein the first data is that the vehicle is driving on a flat road or is stationary, and the radar detects strong reflection road guardrail data meeting a preset condition; the second data is self-calibration program start control data; generating and obtaining third data corresponding to the first data, and calculating fourth data corresponding to the third data, creating and determining a first change rule corresponding to the fourth data; wherein the third data is distance data and original pitch angle data of multiple reflection points on the guardrail; the fourth data is height value data of each reflection point; the first change rule includes: a rule that the height value gradually increases from near to far, a rule that the height value gradually decreases from near to far, and a rule that the height value is basically unchanged; based on the first change rule, combined with a first preset threshold, iteratively compensating and generating fifth data corresponding to the real-time state of the vehicle, and calculating sixth data corresponding to the fifth data; wherein the first preset threshold is a preset step data; the fifth data is the pitch angle data after compensation processing; the sixth data is the height value data of each reflection point after compensation processing; generating and obtaining seventh data corresponding to the first change rule, and generating and outputting eighth data corresponding to the seventh data and used for subsequent angle measurement correction processing of the radar; wherein the seventh data is pitch angle deviation compensation value data.

2. The method according to claim 1, wherein, The judgment basis for the strong reflection road guardrail data meeting the preset condition includes that the signal-to-noise ratio of the guardrail point cloud exceeds a first threshold, the height estimation value of the guardrail exceeds a second threshold, and the distance length of the continuous appearance of the guardrail exceeds a third threshold; wherein the first threshold is a signal-to-noise ratio threshold; the second threshold is a guardrail height threshold; and the third threshold is a guardrail continuous length threshold.

3. The method according to claim 1 or 2, characterized in that, The judgment basis for the flat road comes from the attitude angular velocity information output by the vehicle inertial navigation unit; wherein when the angular velocity is lower than a set threshold, it is determined that the current road section is a flat road section.

4. The method of claim 1, wherein, Based on the first change rule, combined with a first preset threshold, iteratively compensating and generating fifth data corresponding to the real-time state of the vehicle, and calculating sixth data corresponding to the fifth data, further comprises: if the initial judged height change conforms to the rule of height value gradually decreasing from near to far in the first change rule, then positive step compensation processing is performed until the first change rule changes to the rule of height value gradually increasing from near to far; if the initial judged height change conforms to the rule of height value gradually increasing from near to far in the first change rule, then negative step compensation processing is performed until the change rule changes to the rule of height value gradually decreasing from near to far.

5. The road fence based 4D imaging millimeter wave radar pitch angle dynamic self-calibration processing method according to claim 4, characterized in that, The preset step length of the step compensation is 0.1°; wherein the preset cumulative compensation angle range is ±3°.

6. The method of claim 1 or 4, wherein, The iterative compensation processing adopts a variable step strategy for iterative compensation processing; wherein a first larger step length is initially used for compensation, and when the change rule is reversed, a second smaller step length is used for reverse or fine compensation processing.

7. The method of claim 1, wherein, The generating and acquiring a seventh data corresponding to the first change rule, and generating and outputting a corresponding eighth data for subsequent radar angle measurement correction processing based on the seventh data are further included. Using the determined cumulative compensation value data, the guardrail height value data of multiple preset position points at the far field distance is calculated and generated; It is judged whether the height difference of the corresponding far field position point is within the allowed tolerance range, if yes, it is confirmed that the calibration is successful, otherwise, the cumulative compensation value is fine-tuned and verified again.

8. A road fence based 4D imaging millimeter wave radar pitch angle dynamic self-calibration processing system, characterized in that, The system is used to implement the 4D imaging millimeter wave radar pitch angle dynamic self-calibration processing method based on the road guardrail according to any one of claims 1 to 7, and the system comprises: A first data generation and acquisition unit is configured to generate and acquire first data corresponding to the real-time state of the vehicle, and generate corresponding second data based on the first data. The first data is strong reflection road guardrail data detected by the radar when the vehicle is driving or stationary on a flat road and meets a preset condition. The second data is self-calibration program start control data. A second data generation and acquisition unit is configured to generate and acquire third data corresponding to the first data, calculate and generate fourth data corresponding to the third data, and create and determine a first change rule corresponding to the fourth data. The third data is distance data and original pitch angle data of multiple reflection points on the guardrail. The fourth data is height value data of each reflection point. The first change rule includes a rule that the height value gradually increases from near to far, a rule that the height value gradually decreases from near to far, and a rule that the height value is basically unchanged. A data iterative compensation processing unit is configured to iteratively compensate and process the fifth data corresponding to the real-time state of the vehicle based on the first change rule and a first preset threshold, and calculate and generate the sixth data corresponding to the fifth data. The first preset threshold is a preset step length data. The fifth data is the pitch angle data after compensation processing. The sixth data is the height value data of each reflection point after compensation processing. A data generation and output unit is configured to generate and acquire seventh data corresponding to the first change rule, and generate and output corresponding eighth data for subsequent radar angle measurement correction processing based on the seventh data. The seventh data is the pitch angle deviation compensation value data.

9. A road fence based 4D imaging millimeter wave radar pitch angle dynamic self-calibration processing system according to claim 8, characterized in that, The judgment basis for the strong reflection road guardrail data meeting the preset condition includes that the signal-to-noise ratio of the guardrail point cloud exceeds a first threshold, the height estimate value of the guardrail exceeds a second threshold, and the distance length of the continuous appearance of the guardrail exceeds a third threshold. The first threshold is a signal-to-noise ratio threshold. The second threshold is a guardrail height threshold. The third threshold is a guardrail continuous length threshold. The judgment basis of the flat road comes from the attitude angular velocity information output by the vehicle inertial navigation unit. When the angular velocity is lower than the set threshold, it is determined that the current road section is a flat road section. The data iterative compensation processing unit further comprises: The first determination module is used to perform positive step compensation processing if the initial determination of the height change conforms to the rule that the height value gradually decreases from near to far in the first change rule, until the first change rule changes to the rule that the height value gradually increases from near to far. The second determination module is used to perform negative step compensation processing if the initial determination of the height change conforms to the first change law where the height value gradually increases from near to far, until the change law changes to the law where the height value gradually decreases from near to far. The preset step size for the step compensation is 0.1°; wherein the preset cumulative compensation angle range is ±3°. The iterative compensation process employs a variable step size strategy; initially, a larger step size is used for compensation, and when the change pattern reverses, a smaller step size is used for reverse or fine compensation. The system also includes: The first generation module is used to calculate and generate guardrail height data at multiple preset location points at far-field distances using determined cumulative compensation value data; The third judgment module is used to determine whether the height difference of the corresponding far-field position point is within the allowable tolerance range. If so, the calibration is confirmed to be successful; otherwise, the cumulative compensation value is fine-tuned and then verified again.

10. A road fence based 4D imaging millimeter wave radar pitch angle dynamic self-calibration processing platform, characterized in that, The system includes a processor, a memory, and a control program for a 4D imaging millimeter-wave radar elevation angle dynamic self-calibration processing platform based on road guardrails. The processor executes the control program, which is stored in the memory. This control program implements the 4D imaging millimeter-wave radar elevation angle dynamic self-calibration processing method based on road guardrails, as described in any one of claims 1 to 7.

Images