Millimeter wave radar multi-mode azimuth angle relative deviation automatic correction method and system

Abstract

The invention discloses a millimeter wave radar multi-mode azimuth angle relative deviation automatic correction method and system. The method comprises the following steps: determining a reference mode, a to-be-corrected mode and a common detection area of the reference mode and the to-be-corrected mode; determining a statistical area based on the public detection area, performing grid division on the statistical area, and counting effective detection points of the reference mode and the mode to be corrected in each grid; comparing the sum of the multi-frame accumulation results of the number of the effective detection points in all the grids in the statistical area in the reference mode and the mode to be corrected with a preset threshold value, and determining the azimuth angle deviation of the mode to be corrected relative to the reference mode according to the comparison result, and correcting the azimuth angle of the detection point in the to-be-corrected mode. According to the method, the purpose of compensating the angle measurement relative deviation between different modes can be achieved, so that the trace point continuity between the different modes is improved, and the overall quality of the track is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of millimeter wave radars, and in particular to a method and system for automatically correcting relative deviations of multi-mode azimuth angles of millimeter wave radars. Background Art

[0002] Millimeter-wave radar plays an extremely important role in many fields such as automobile safety, autonomous driving, and smart transportation. In practice, in order to achieve blind-spot coverage of multiple distance segments, millimeter-wave radar systems usually work in combination with multiple modes, each of which covers a different distance segment. In addition, in order to achieve blind-spot coverage, the distance segments covered by different modes will overlap, which is called the common detection area.

[0003] However, since different modes usually use different antenna boards or different transceiver modes, and different antenna boards or transceiver modes correspond to different antenna parameter configurations, amplitude and phase factors, hardware performance, etc., there will inevitably be relative deviations in the angle measurement results between different modes, which is not conducive to the fusion and tracking of detection information between different modes, thereby affecting the track formation. Summary of the invention

[0004] The present application provides a method and system for automatically correcting relative deviation of multi-mode azimuth angles of a millimeter-wave radar, so as to solve the problem of relative deviation of angle measurement results between different modes of the existing millimeter-wave radar system.

[0005] According to the first aspect, an embodiment provides a method for automatically correcting relative deviation of multi-mode azimuth angles of a millimeter-wave radar, the method comprising:

[0006] Determining a reference pattern and a pattern to be corrected and a common detection area of ​​the reference pattern and the pattern to be corrected;

[0007] Determine a statistical area based on the common detection area, divide the statistical area into grids, and count the number of valid detection points of the reference mode and the mode to be corrected in each grid;

[0008] Compare the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the mode to be corrected respectively with a preset threshold, and determine the azimuth angle deviation of the mode to be corrected relative to the reference mode according to the comparison result;

[0009] Based on the azimuth angle deviation, the azimuth angle of the detection point in the to-be-corrected mode is corrected.

[0010] Further, determining the reference pattern and the pattern to be corrected and a common detection area of ​​the reference pattern and the pattern to be corrected specifically includes:

[0011] A mode with relatively high angle measurement accuracy is selected from at least two modes having a common detection area as a reference mode, and a mode with relatively low angle measurement accuracy is selected as a mode to be corrected.

[0012] Further, determining a statistical area based on the common detection area, dividing the statistical area into grids, and counting the number of valid detection points of the reference mode and the mode to be corrected in each grid, specifically includes:

[0013] A statistical area is selected from the common detection area, and the horizontal coordinate range of the statistical area is X min ~X max , the vertical axis range is Y min ~Y max ;

[0014] In a Cartesian rectangular coordinate system formed with the radar as the origin, the two dimensions of the horizontal coordinate range and the vertical coordinate range of the statistical area are evenly divided into K and L grids respectively, or in a polar coordinate system formed with the radar as the origin, the two dimensions of the radial distance coordinate range and the azimuth angle coordinate range of the statistical area are evenly divided into K and L grids respectively; the dimension divided into K grids is used as the first coordinate dimension, and the dimension divided into L grids is used as the second coordinate dimension;

[0015] Detection points whose signal-to-noise ratio in each frame falls within the statistical area and is greater than a preset threshold value and whose radial velocity exceeds a preset value are selected as valid detection points.

[0016] Furthermore, the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the mode to be corrected is compared with a preset threshold, and the azimuth angle deviation of the mode to be corrected relative to the reference mode is determined according to the comparison result, specifically including:

[0017] In the reference mode and the mode to be corrected, the number of valid detection points of each grid in each frame falling into the statistical area is counted and accumulated over multiple frames to obtain heat map matrices A and B. The heat map matrices A and B need to be initialized to matrices of size K×L with all element values ​​0 before starting the statistics. When each frame is counted, for the kth grid in the first coordinate dimension and the lth grid in the second coordinate dimension, if the number of valid detection points of the current frame falling into the corresponding grid in the reference mode and the mode to be corrected is h1 and h2 respectively, then A k,l The value of h1 increases, B k,l The value of increases by h2;

[0018] The sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the correction mode is respectively compared with the preset minimum threshold N min and the maximum threshold Nmax Compare and calculate or update the azimuth angle deviation θ of the mode to be corrected relative to the reference mode based on the comparison result bias .

[0019] Further, based on the comparison result, the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is calculated or updated. bias , specifically including:

[0020] If the sum of the multi-frame accumulation results of the number of valid detection points in all grids in the statistical area in any of the modes is less than N min , then the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is not estimated bias , at this time θ bias =0.

[0021] Further, based on the comparison result, the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is calculated or updated. bias , specifically including:

[0022] If the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in each mode is not less than N min , but are all smaller than N max , then the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is updated according to the thermal map matrices A and B of the reference mode and the mode to be corrected bias , specifically including:

[0023] The heat map matrix B of the to-be-corrected mode is sequentially shifted by i=-M, -M+1, ..., -1, 0, 1, ..., M-1, M grids along the dimension corresponding to the horizontal coordinate range or the azimuth angle coordinate range of the statistical area to obtain a 2M+1 matrix after shifting:

[0024] C i (i=-M,-M+1,...,-1,0,1,...,M-1,M)

[0025] Wherein, M represents the number of grids corresponding to the preset maximum azimuth angle deviation of the mode to be corrected relative to the reference mode, i takes a negative sign to indicate a direction deviation corresponding to a decrease in the azimuth angle, and i takes a positive sign to indicate a direction deviation corresponding to an increase in the azimuth angle;

[0026] Calculate the 2M+1 matrices C of the thermal map matrix B of the mode to be corrected after the offset i The mutual correlation coefficient of (i=-M,-M+1,...,-1,0,1,...,M-1,M) and the thermal map matrix A of the reference mode is:

[0027]

[0028] Wherein i=-M,-M+1,...,-1,0,1,...,M-1,M, l=1,2,...,L, k=1,2,...,K, L and K respectively represent the number of rows and columns of the grid after the statistical area is equally divided into a plurality of grids, and finally 2M+1 mutual correlation coefficients are obtained;

[0029] From the 2M+1 mutual correlation coefficient α i Find the grid number offset i corresponding to the largest value m , according to i m Further calculate and update the azimuth angle offset θ of the mode to be corrected relative to the reference mode bias :

[0030] For the Cartesian coordinate system,

[0031]

[0032] Among them, Δx represents the horizontal distance spanned by each grid when the horizontal coordinate range of the statistical area is evenly divided into multiple grids;

[0033] For polar coordinates,

[0034] θ bias =i m ·Δθ

[0035] Wherein, Δθ represents the azimuth angle spanned by each grid when the azimuth angle coordinate range of the statistical area is evenly divided into multiple grids.

[0036] Further, based on the comparison result, the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is calculated or updated. bias , specifically including:

[0037] If the sum of the multi-frame accumulation results of the number of valid detection points in all grids in the statistical area in any of the modes is not less than N max , then the number of valid detection points in all grids in the statistical area under the corresponding mode reaches N max The azimuth angle deviation θ calculated in the last frame bias As the azimuth angle deviation of the pattern to be corrected relative to the reference pattern, θ is used in all frames after the last frame. bias As the azimuth angle deviation of the pattern to be corrected relative to the reference pattern.

[0038] Further, based on the azimuth angle deviation, the azimuth angle of the detection point in the correction mode is corrected, specifically including:

[0039] According to the offset θ biasCorrect the azimuth angles of all detection points in the correction mode; assume that the original angle estimation result of the pth detection point to be corrected is θ p (p=1,2,...,P), then the corrected angle dimension θ c =θ p +θ bias , where P is the number of detection points in the mode to be corrected.

[0040] Furthermore, the method further comprises:

[0041] If the radar has P (P>2) modes, where mode 1 and mode 2 have a common detection area, mode 2 and mode 3 have a common detection area, and so on, mode P and mode P-1 have a common detection area, then first take mode 1 as the reference mode and calculate the azimuth angle deviation θ of mode 2 relative to mode 1 bias21 , and then take mode 2 as the reference mode to calculate the azimuth angle deviation θ of mode 3 relative to mode 2 bias32 , and so on, until the azimuth angle deviation of mode P relative to mode P-1 is calculated, thereby achieving azimuth angle correction for each mode.

[0042] According to the second aspect, an embodiment provides a millimeter wave radar multi-mode azimuth angle relative deviation automatic correction system, the system comprising:

[0043] A reference mode determination module, used to determine a reference mode and a mode to be corrected and a common detection area of ​​the reference mode and the mode to be corrected;

[0044] A statistical module, used to determine a statistical area based on the common detection area, divide the statistical area into grids, and count the number of valid detection points of the reference mode and the mode to be corrected in each grid;

[0045] An angle deviation determination module, used to compare the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the mode to be corrected respectively with a preset threshold, and determine the azimuth angle deviation of the mode to be corrected relative to the reference mode according to the comparison result;

[0046] The correction module is used to correct the azimuth angle of the detection point in the correction mode based on the azimuth angle deviation.

[0047] The present application provides a method and system for automatically correcting the relative deviation of the azimuth angle of a millimeter-wave radar multi-mode, which determines a reference mode and a mode to be corrected and a common detection area of ​​the reference mode and the mode to be corrected; determines a statistical area based on the common detection area, divides the statistical area into grids, and counts the number of valid detection points of the reference mode and the mode to be corrected in each grid; compares the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area under the reference mode and the mode to be corrected respectively with a preset threshold, and determines the azimuth angle deviation of the mode to be corrected relative to the reference mode according to the comparison result; based on the azimuth angle deviation, corrects the azimuth angle of the detection point in the mode to be corrected. The method of the present invention is applicable to a millimeter-wave radar including multiple modes and a common detection area between adjacent modes, so as to achieve the purpose of compensating for the relative deviation of angle measurement between different modes, thereby improving the continuity of point tracks between different modes and effectively improving the overall quality of the track. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flowchart of a method for automatically correcting relative deviation of multi-mode azimuth angles of a millimeter-wave radar provided by an embodiment of the present invention;

[0049] Figure 2 A schematic diagram of detection areas covered by different modes and a common detection area in an XY coordinate system in a method for automatic correction of relative deviation of multi-mode azimuth angles of a millimeter-wave radar provided by an embodiment of the present invention;

[0050] Figure 3 A schematic diagram of dividing a common detection area into grids in an R-θ coordinate system in a method for automatically correcting relative deviations of multi-mode azimuth angles of a millimeter-wave radar provided by an embodiment of the present invention;

[0051] Figure 4 A comparison diagram of different mode traces in actual scenes before and after azimuth angle correction in a method for automatically correcting relative deviation of multi-mode azimuth angles of a millimeter-wave radar provided by an embodiment of the present invention;

[0052] Figure 5 A schematic diagram of the logical structure of a millimeter-wave radar multi-mode azimuth angle relative deviation automatic correction system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present application better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present application are not shown or described in the specification, this is to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0054] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0055] The first embodiment of the present invention provides a method for automatically correcting relative deviation of multi-mode azimuth angles of millimeter wave radars. Figure 1 Provide detailed explanation.

[0056] In this embodiment, it is assumed that the radar has two modes, and the relative deviation of the azimuth angles between the two modes is automatically corrected.

[0057] The different modes referred to in this embodiment refer to working modes responsible for detecting different distance segments; in addition to the difference in distance segments, different modes generally use different angle measurement arrays, and the corresponding angle resolution, angle measurement accuracy, etc. may also be different. Because of this, there will be relative deviations in the angle measurement results between different modes, which need to be corrected; there is no restriction on whether different modes work at the same time. From a macroscopic point of view, different modes work at the same time, so as to achieve simultaneous coverage of all distance segments, but from a microscopic point of view, it is impossible to achieve absolute "simultaneity". For example, the same frame may need to transmit the signal of mode 1 and then the signal of mode 2. The two cannot be transmitted at the same time, but can only be transmitted alternately, but the alternation time is in microseconds. Therefore, from a macroscopic point of view, it can be understood that the two modes work at the same time.

[0058] like Figure 1 As shown, in step S100, a reference pattern and a pattern to be corrected and a common detection area of ​​the reference pattern and the pattern to be corrected are determined.

[0059] The above steps specifically include:

[0060] Step S110, determining a reference mode and a mode to be corrected according to the angle measurement accuracy of different modes. In this example, a mode with higher angle measurement accuracy is selected as the reference mode, and the other mode is the mode to be corrected.

[0061] Step S120, determining the common detection area covered by the reference pattern and the pattern to be corrected. Figure 2 As shown, the detection area covered by mode 1 and the detection area covered by mode 2 may overlap, that is, there is a common detection area.

[0062] like Figure 1 As shown, in step S200, a statistical area is determined based on the common detection area, the statistical area is divided into grids, and the number of valid detection points of the reference mode and the mode to be corrected in each grid is counted.

[0063] The above steps specifically include:

[0064] Firstly, all detection points in the common detection area and the parameter estimation results of each detection point are extracted from the detection results obtained by processing each frame of echo signals of the reference mode and the mode to be corrected.

[0065] It should be noted that after the radar transmits the signal and the echo signal is reflected by the target (such as pedestrians and vehicles), the radar signal processor will process the echo signal and detect the target points such as pedestrians and vehicles, which are called detection points. The radar signal processor will also perform parameter estimation on each detection point to obtain the distance, speed, azimuth angle, horizontal coordinate, vertical coordinate, signal-to-noise ratio and other information of each detection point.

[0066] Step S210, the echo signals received in each mode of each frame are processed separately, and the signal processing methods include but are not limited to existing technologies such as distance dimension FFT (fast Fourier transform), Doppler dimension FFT, non-coherent accumulation, CFAR (constant false alarm) detection, azimuth angle dimension FFT, etc., to obtain all target detection points detected in each mode of each frame and parameter estimation results of each target detection point, including distance, speed, azimuth angle, horizontal coordinate, vertical coordinate, signal-to-noise ratio and other information.

[0067] Step S220, screening all detection points falling into the common detection area from all detection points of each mode in each frame and parameter estimation results of each detection point, including distance, speed, azimuth angle, horizontal coordinate, vertical coordinate, signal-to-noise ratio and other information.

[0068] After obtaining all detection points that fall into the common detection area of ​​each frame reference pattern and the pattern that needs to be corrected and the parameter estimation results of each detection point, the azimuth angle deviation of the pattern that needs to be corrected relative to the reference pattern is determined based on the reference pattern and the detection points of multiple frames of the pattern that needs to be corrected in the common area and the parameter estimation results of each detection point, and the azimuth angle correction is performed.

[0069] Step S230, selecting a statistical area from the common detection area of ​​the reference mode and the mode to be corrected, the horizontal coordinate range of the statistical area is X min ~X max , the vertical axis range is Y min ~Y max .

[0070] A statistical area is selected from the common detection area of ​​the reference mode and the mode to be corrected. The statistical area may be the entire common detection area or an area in the common detection area where the probability of the target point appearing is high.

[0071] Step S240, the statistical area is equally divided into a plurality of grids. In a Cartesian rectangular coordinate system formed with the radar as the origin, the horizontal coordinate range and the vertical coordinate range of the statistical area are equally divided into K and L grids respectively, wherein the horizontal length of each grid is X step =(X max -X min ) / K, the vertical length of each grid is Y step =(Y max -Y min ) / L, or in a polar coordinate system formed with the radar as the origin, the two dimensions of the radial distance coordinate range and the azimuth angle coordinate range of the statistical area can be evenly divided into K and L grids respectively. Figure 3 This is a schematic diagram of dividing the public detection area into grids in the R-θ coordinate system formed by the radar origin. The azimuth coordinate range of the statistical area is θ min ~θ max , the radial distance coordinate range is R min ~R max .

[0072] Step S250 , screening detection points in each frame reference mode and the mode to be corrected, whose signal-to-noise ratio is greater than the 12dB threshold and whose absolute value of radial velocity is greater than 2m / s and falls within the statistical area, as valid detection points.

[0073] like Figure 1 As shown, in step S300, the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the mode to be corrected is compared with a preset threshold, and the azimuth angle deviation of the mode to be corrected relative to the reference mode is determined based on the comparison result.

[0074] The above steps specifically include:

[0075] Step S310, in the reference mode and the mode to be corrected, the number of valid detection points of each frame falling into each grid in the statistical area is counted and accumulated over multiple frames, and the heat map matrices A and B are obtained. The heat map matrices A and B need to be initialized to matrices of size K×L and element values ​​​​are all 0 before starting the statistics. When each frame is counted, for the kth grid under the first coordinate dimension and the lth grid under the second coordinate dimension, if the number of valid detection points of the current frame falling into the corresponding grid in the reference mode and the mode to be corrected is h1 and h2 respectively, then A k,l The value of h1 increases, B k,l The value of h2 is increased.

[0076] Step S320, the sum of the multi-frame accumulation results of the number of valid detection points falling into all grids in the statistical area in the reference mode and the to-be-corrected mode is compared with the preset minimum threshold N min and the maximum threshold N max (In this implementation, the minimum threshold N of the number of valid detection points accumulated in multiple frames in a single mode min Should be set large enough, the maximum threshold N max The azimuth angle deviation θ of the mode to be corrected relative to the reference mode is calculated or updated based on the comparison result. bias There are three specific cases:

[0077] (1) If the sum of the multi-frame accumulation results of the number of valid detection points in all grids in the statistical area in any mode is less than N min , then the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is not estimated bias , at this time θ bias =0.

[0078] (2) If the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in each mode is not less than N min , but are all smaller than N max , then the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is updated according to the thermal map matrices A and B of the reference mode and the mode to be corrected bias , specifically including the following steps:

[0079] The heat map matrix B of the mode to be corrected is shifted i=-M, -M+1, ..., -1, 0, 1, ..., M-1, M grids in sequence along the dimension corresponding to the horizontal coordinate range (Cartesian coordinate system) or the azimuth angle coordinate range (polar coordinate system) of the statistical area to obtain the shifted 2M+1 matrix:

[0080] C i (i=-M,-M+1,...,-1,0,1,...,M-1,M)

[0081] Where M represents the maximum possible azimuth angle deviation Δθ of the mode to be corrected relative to the reference mode max (It is known that the azimuth angle deviation between the two modes cannot exceed Δθ max ) The number of grids corresponding to the azimuth angle is negative, indicating that the corresponding direction is offset when the azimuth angle decreases, and the positive sign of i indicates that the corresponding direction is offset when the azimuth angle increases;

[0082] In this embodiment, in the XY coordinate system, M can be determined according to the following formula:

[0083]

[0084] Then calculate the 2M+1 matrix C after the offset of the thermal map matrix B of the mode to be corrected. i The mutual correlation coefficient of (i=-M,-M+1,...,-1,0,1,...,M-1,M) and the thermal map matrix A of the reference mode is:

[0085]

[0086] Wherein i=-M,-M+1,...,-1,0,1,...,M-1,M, l=1,2,...,L, k=1,2,...,K, L and K respectively represent the number of rows and columns of the grid after the statistical area is equally divided into a plurality of grids, and finally 2M+1 mutual correlation coefficients are obtained;

[0087] From the 2M+1 mutual correlation coefficient α i Find the grid number offset i corresponding to the largest value m , according to i m Further calculate and update the azimuth angle offset θ of the mode to be corrected relative to the reference mode bias :

[0088] For the Cartesian coordinate system,

[0089]

[0090] Wherein, Δx represents the horizontal distance spanned by each grid when the horizontal coordinate range of the statistical area is evenly divided into multiple grids;

[0091] For polar coordinates,

[0092] θ bias =i m ·Δθ

[0093] Wherein, Δθ represents the azimuth angle spanned by each grid when the azimuth angle coordinate range of the statistical area is evenly divided into multiple grids.

[0094] (3) If the sum of the multi-frame accumulation results of the number of valid detection points in all grids in the statistical area in any of the modes is not less than N max , then the number of valid detection points in all grids in the statistical area under the corresponding mode reaches N max The azimuth angle deviation θ calculated in the last frame bias As the azimuth angle deviation of the pattern to be corrected relative to the reference pattern, θ is used in all frames after the last frame. bias As the azimuth angle deviation of the mode to be corrected relative to the reference mode, that is, the error θ bias The value of is updated.

[0095] like Figure 1 As shown, in step S400, based on the azimuth angle deviation, the azimuth angle of the detection point in the correction mode is corrected.

[0096] Specifically, according to the offset θ bias Correct the azimuth angles of all detection points in the correction mode; assume that the original angle estimation result of the pth detection point to be corrected is θ p (p=1,2,...,P), then the corrected angle dimension θ c =θ p +θ bias , where P is the number of detection points in the mode to be corrected.

[0097] Attached Figure 4 Attached is a comparison diagram of different pattern traces in actual scenes before and after azimuth angle correction using the technical solution of the present invention. Figure 4 (a) shows the dot trace diagram without using the technical solution of the present invention for azimuth angle calibration, in which the black and gray dot traces represent the two modes responsible for different distance segments, respectively. Due to the azimuth angle deviation between the two modes, it is reflected in the xy (x represents the horizontal distance, y represents the vertical distance) dot trace diagram as a relative deviation in the x direction between the two modes in the common detection area, as shown in the boxed part. After using the technical solution of the present invention for azimuth angle calibration, the azimuth deviation is effectively corrected, as shown in the attached figure. Figure 4 (b) The box marked part is shown.

[0098] In another embodiment of the present invention, if the radar has P (P>2) modes, where mode 1 and mode 2 have a common detection area, mode 2 and mode 3 have a common detection area, and so on, mode P and mode P-1 have a common detection area, then firstly take mode 1 as the reference mode, and calculate the azimuth angle deviation θ of mode 2 relative to mode 1 bias21 , and then take mode 2 as the reference mode to calculate the azimuth angle deviation θ of mode 3 relative to mode 2 bias32 , and so on, until the azimuth angle deviation of mode P relative to mode P-1 is calculated, thereby achieving azimuth angle correction of each mode. The specific steps can refer to the above embodiment, which will not be repeated here.

[0099] Corresponding to the above-disclosed method for automatically correcting relative deviation of multi-mode azimuth angles of millimeter-wave radar, the embodiment of the present invention further discloses a system for automatically correcting relative deviation of multi-mode azimuth angles of millimeter-wave radar, such as Figure 5 As shown, it specifically includes:

[0100] A reference mode determination module, used to determine a reference mode and a mode to be corrected and a common detection area of ​​the reference mode and the mode to be corrected;

[0101] A statistical module, used to determine a statistical area based on the common detection area, divide the statistical area into grids, and count the number of valid detection points of the reference mode and the mode to be corrected in each grid;

[0102] An angle deviation determination module, used to compare the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the mode to be corrected respectively with a preset threshold, and determine the azimuth angle deviation of the mode to be corrected relative to the reference mode according to the comparison result;

[0103] The correction module is used to correct the azimuth angle of the detection point in the correction mode based on the azimuth angle deviation.

[0104] It should be noted that for the detailed description of a millimeter-wave radar multi-mode azimuth angle relative deviation automatic correction system provided in an embodiment of the present invention, reference can be made to the relevant description of a millimeter-wave radar multi-mode azimuth angle relative deviation automatic correction method provided in an embodiment of the present application, which will not be repeated here.

[0105] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art, according to the concept of the present invention, some simple deductions, modifications or substitutions can be made.

Claims

1. A method for automatically correcting relative deviation of multi-mode azimuth angle of millimeter wave radar, characterized in that: The method comprises: Determining a reference pattern and a pattern to be corrected and a common detection area of ​​the reference pattern and the pattern to be corrected; Determine a statistical area based on the common detection area, divide the statistical area into grids, and count the number of valid detection points of the reference mode and the mode to be corrected in each grid; Compare the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the mode to be corrected respectively with a preset threshold, and determine the azimuth angle deviation of the mode to be corrected relative to the reference mode according to the comparison result; Based on the azimuth angle deviation, the azimuth angle of the detection point in the to-be-corrected mode is corrected.

2. A millimeter wave radar multi-mode azimuth angle relative deviation automatic correction method as claimed in claim 1, characterized in that: Determining a reference pattern and a pattern to be corrected and a common detection area of ​​the reference pattern and the pattern to be corrected specifically includes: A mode with relatively high angle measurement accuracy is selected from at least two modes having a common detection area as a reference mode, and a mode with relatively low angle measurement accuracy is selected as a mode to be corrected.

3. The method for automatically correcting relative deviation of multi-mode azimuth angle of millimeter wave radar according to claim 1, characterized in that: Determining a statistical area based on the common detection area, dividing the statistical area into grids, and counting the number of valid detection points of the reference mode and the mode to be corrected in each grid, specifically includes: A statistical area is selected from the public detection area, and the horizontal coordinate range of the statistical area is X min ~X max , the vertical axis range is Y min ~Y max ; In a Cartesian rectangular coordinate system formed with the radar as the origin, the two dimensions of the horizontal coordinate range and the vertical coordinate range of the statistical area are evenly divided into K and L grids respectively, or in a polar coordinate system formed with the radar as the origin, the two dimensions of the radial distance coordinate range and the azimuth angle coordinate range of the statistical area are evenly divided into K and L grids respectively; the dimension divided into K grids is used as the first coordinate dimension, and the dimension divided into L grids is used as the second coordinate dimension; Detection points whose signal-to-noise ratio in each frame falls within the statistical area and is greater than a preset threshold value and whose radial velocity exceeds a preset value are selected as valid detection points.

4. A millimeter wave radar multi-mode azimuth angle relative deviation automatic correction method as claimed in claim 3, characterized in that: The sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the mode to be corrected is compared with a preset threshold, and the azimuth angle deviation of the mode to be corrected relative to the reference mode is determined according to the comparison result, specifically including: In the reference mode and the mode to be corrected, the number of valid detection points of each grid in each frame falling into the statistical area is counted and accumulated over multiple frames to obtain heat map matrices A and B. The heat map matrices A and B need to be initialized to matrices of size K×L and element values ​​of all 0 before starting the statistics. When each frame is counted, for the kth grid in the first coordinate dimension and the lth grid in the second coordinate dimension, if the number of valid detection points of the current frame falling into the corresponding grid in the reference mode and the mode to be corrected is h1 and h2 respectively, then the heat map matrix A of the kth grid in the first coordinate dimension and the lth grid in the second coordinate dimension in the reference mode is k,l The value of is increased by h1, and the heat map matrix B of the kth grid in the first coordinate dimension and the lth grid in the second coordinate dimension in the correction mode is k,l The value of increases by h2; The sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the correction mode is respectively compared with the preset minimum threshold N min and the maximum threshold N max Compare and calculate or update the azimuth angle deviation θ of the mode to be corrected relative to the reference mode based on the comparison result bias .

5. A millimeter wave radar multi-mode azimuth angle relative deviation automatic correction method as claimed in claim 4, characterized in that: Based on the comparison result, the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is calculated or updated. bias , specifically including: If the sum of the multi-frame accumulation results of the number of valid detection points in all grids in the statistical area in any of the modes is less than N min , then the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is not estimated bias , at this time θ bias =0.

6. A millimeter wave radar multi-mode azimuth angle relative deviation automatic correction method as claimed in claim 5, characterized in that: Based on the comparison result, the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is calculated or updated. bias , specifically including: If the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in each mode is not less than N min , but are all smaller than N max , then the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is updated according to the thermal map matrices A and B of the reference mode and the mode to be corrected bias , specifically including: The heat map matrix B of the to-be-corrected mode is sequentially shifted by i=-M, -M+1, ..., -1, 0, 1, ..., M-1, M grids along the dimension corresponding to the horizontal coordinate range or the azimuth angle coordinate range of the statistical area to obtain a 2M+1 matrix after shifting: C i (i=-M,-M+1,...,-1,0,1,...,M-1,M) Wherein, M represents the number of grids corresponding to the preset maximum azimuth angle deviation of the mode to be corrected relative to the reference mode, i takes a negative sign to indicate a direction deviation corresponding to a decrease in the azimuth angle, and i takes a positive sign to indicate a direction deviation corresponding to an increase in the azimuth angle; Calculate the 2M+1 matrices C of the thermal map matrix B of the mode to be corrected after the offset i The mutual correlation coefficient of (i=-M,-M+1,...,-1,0,1,...,M-1,M) and the thermal map matrix A of the reference mode is: Wherein i=-M,-M+1,...,-1,0,1,...,M-1,M, l=1,2,...,L, k=1,2,...,K, L and K respectively represent the number of rows and columns of the grid after the statistical area is equally divided into a plurality of grids, and finally 2M+1 mutual correlation coefficients are obtained; From the 2M+1 mutual correlation coefficient α i Find the grid number offset i corresponding to the largest value m , according to i m Further calculate and update the azimuth angle offset θ of the mode to be corrected relative to the reference mode bias : For the Cartesian coordinate system, Among them, Δx represents the horizontal distance spanned by each grid when the horizontal coordinate range of the statistical area is evenly divided into multiple grids; For polar coordinates, i bias =i m ·Dth Wherein, Δθ represents the azimuth angle spanned by each grid when the azimuth angle coordinate range of the statistical area is evenly divided into multiple grids.

7. A millimeter wave radar multi-mode azimuth angle relative deviation automatic correction method as claimed in claim 6, characterized in that: Based on the comparison result, the azimuth angle deviation θ of the mode to be corrected relative to the reference mode is calculated or updated. bias , specifically including: If the sum of the multi-frame accumulation results of the number of valid detection points in all grids in the statistical area in any of the modes is not less than N max , then the number of valid detection points in all grids in the statistical area under the corresponding mode reaches N max The azimuth angle deviation θ calculated in the last frame bias As the azimuth angle deviation of the pattern to be corrected relative to the reference pattern, θ is used in all frames after the last frame. bias As the azimuth angle deviation of the pattern to be corrected relative to the reference pattern.

8. A millimeter wave radar multi-mode azimuth angle relative deviation automatic correction method as claimed in claim 7, characterized in that: Based on the azimuth angle deviation, the azimuth angle of the detection point in the correction mode is corrected, specifically including: According to the offset θ bias Correct the azimuth angles of all detection points in the correction mode; assume that the original angle estimation result of the pth detection point to be corrected is θ p (p=1,2,...,P), then the corrected angle dimension θ c =θ p +θ bias , where P is the number of detection points in the mode to be corrected.

9. The method for automatically correcting relative deviation of multi-mode azimuth angle of millimeter wave radar according to claim 1, characterized in that: The method further comprises: If the radar has P (P>2) modes, where mode 1 and mode 2 have a common detection area, mode 2 and mode 3 have a common detection area, and so on, mode P and mode P-1 have a common detection area, then first take mode 1 as the reference mode and calculate the azimuth angle deviation θ of mode 2 relative to mode 1 bias21 , and then take mode 2 as the reference mode to calculate the azimuth angle deviation θ of mode 3 relative to mode 2 bias32 , and so on, until the azimuth angle deviation of mode P relative to mode P-1 is calculated, thereby achieving azimuth angle correction for each mode.

10. A millimeter wave radar multi-mode azimuth angle relative deviation automatic correction system, characterized in that: The system comprises: A reference mode determination module, used to determine a reference mode and a mode to be corrected and a common detection area of ​​the reference mode and the mode to be corrected; A statistical module, used to determine a statistical area based on the common detection area, divide the statistical area into grids, and count the number of valid detection points of the reference mode and the mode to be corrected in each grid; An angle deviation determination module, used to compare the sum of the multi-frame cumulative results of the number of valid detection points in all grids in the statistical area in the reference mode and the mode to be corrected respectively with a preset threshold, and determine the azimuth angle deviation of the mode to be corrected relative to the reference mode according to the comparison result; The correction module is used to correct the azimuth angle of the detection point in the correction mode based on the azimuth angle deviation.

Images