Millimeter wave radar measurement method considering various influencing factors

Abstract

The application discloses a millimeter wave radar measurement method considering multiple influence factors, which comprises millimeter wave radar ranging refinement, refining the ranging value in the precision interval based on the fast Fourier transform method and the linear frequency modulation transform; constructing a ranging error refinement compensation model to obtain the accurate ranging value of the target by the radar; performing target rough positioning according to the compensated distance, azimuth angle and pitch angle of the radar and the target; constructing a millimeter wave radar high-precision measurement model considering multiple influence factors by combining the single observation radar pulse measurement precision difference, the real distance and the signal-to-noise ratio comprehensive weighting to obtain millimeter wave radar high-precision measurement data considering multiple influence factors. The application adopts spatial distance intersection to obtain the corner reflector position. The problem of insufficient millimeter wave radar ranging precision is solved, the influence of radar electronic element delay, angle and meteorological factors is eliminated, and the precise observation model is utilized to realize rapid measurement of the orbital geometric state.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of railway engineering surveying, and in particular to a millimeter wave radar measurement method considering multiple influencing factors. BACKGROUND

[0002] High-speed rail operation safety is faced with problems such as deformation of roadbed, bridge and track structure and geological disaster hidden dangers along the line. Under the condition of high-speed train operation, slight deformation of roadbed and bridge will lead to track irregularity, causing train to run at a reduced speed or even affecting train safety. Therefore, track geometry state detection is crucial to high-speed rail safe operation.

[0003] In order to ensure the safe movement of trains, the railway department needs to regularly detect the track geometry state and maintain and repair the track. At present, Beidou positioning technology has become one of the most mature track detection solutions outside the tunnel due to its high precision, all-day and all-weather advantages. However, the demand for positioning and safety detection in the track detection process is not only outside the tunnel, but also inside the tunnel. As a traditional high-precision measuring equipment, the total station integrates inertial navigation technology and provides reference coordinates and inertial navigation system to compensate for dynamic errors, realizing static measurement of track geometry state parameters (such as track gauge, level and height) inside the tunnel. However, the total station equipment still has problems such as high price, low efficiency and inability to effectively protect data security.

[0004] Millimeter wave radar is a radar system that uses millimeter wave frequency band for detection and ranging. It can penetrate obstacles such as water vapor and smoke and is not affected by lighting conditions. It can identify multiple targets simultaneously and has advantages such as long time series, real-time, high precision (sub-millimeter level), non-contact, etc. Compared with hundreds of thousands of measurement robots, millimeter wave radar has the advantage of high cost performance with a lower price (500-2000 yuan). However, most millimeter wave radars are used for deflection monitoring of structures such as bridges, and how to achieve high-precision positioning in complex tunnel environments has not been thoroughly studied. Therefore, there is an urgent need for a millimeter wave radar measurement method that takes into account multiple influencing factors to eliminate the effects of radar electronic component delay, angle and weather factors, construct a precise observation model and achieve rapid measurement of track geometry state. SUMMARY

[0005] Therefore, the purpose of the present application is to provide a millimeter wave radar measurement method considering multiple influencing factors. To solve the problems of low efficiency, high cost and inability to guarantee data security in track geometry state measurement in railway tunnels.

[0006] To achieve the above purpose, the millimeter wave radar measurement method considering multiple influencing factors provided by the present application comprises the following steps:

[0007] S1, obtain original radar data, and finely process the obtained original radar data to obtain an accurate ranging value of a single radar to a target;

[0008] S2, set the azimuth angle and the elevation angle of the radar to the corner reflector to 0; and construct a ranging error compensation model;

[0009] S3, alternately transform the azimuth angle and the elevation angle of the radar to the corner reflector, and construct a millimeter wave radar angle error model;

[0010] S4, compensate the distance obtained by the radar to the corner reflector by using the ranging error model and the angle error model, measure the temperature and humidity near the radar by using a temperature and humidity meter, analyze the relationship between the temperature, the humidity and the compensated distance, and construct a meteorological correction model;

[0011] S5, obtain an initial coordinate according to the original radar data of S1, obtain a corrected accurate distance observation value by using the ranging error compensation model of S2, the millimeter wave radar angle error model of S3 and the meteorological correction model of S4, construct a millimeter wave radar high-precision measurement model considering multiple influence factors according to the influence of single pulse measurement error, distance length and signal-to-noise ratio on measurement, and obtain millimeter wave radar high-precision measurement data considering multiple influence factors by using the millimeter wave radar high-precision measurement model.

[0012] Further preferably, in S1, the fine processing of the obtained original radar data to obtain an accurate ranging value of a single radar to a target comprises:

[0013] S101, distance value coarse extraction, including taking the corner reflector as a target, calculating the frequency of a discrete beat signal in the original radar data by using FFT, selecting the frequency corresponding to the maximum amplitude signal, and converting to obtain a target radial distance;

[0014] S102, spectral refinement, including coarse extraction of the target radial distance, determining the frequency corresponding to the target between the left and right adjacent spectral lines of the maximum spectral line, and using CZT to refine the spectral line to obtain an unambiguous phase of the target;

[0015] S103, ranging value refinement, calculating the phase difference of two ranging according to the unambiguous phase of the target, and converting the phase difference into a corresponding distance change , which is superimposed into the first ranging value to obtain the next epoch accurate ranging value.

[0016] Further preferably, in S2, the azimuth angle and the elevation angle of the radar to the corner reflector are set to 0; and the ranging error compensation model is constructed, comprising:

[0017] ​Adjust the pitch angle and the horizontal angle of the radar antenna and the corner reflector to 0°, displace the corner reflector from the radar by 0.5 m, with a step of 0.1 m, in a straight line for 2000 times, displace to 200 m, obtain the real absolute distance between the radar and the corner reflector as a true value, and subtract the distance measured by the radar from the corner reflector to obtain an optimal linear function model of the distance true value, as a ranging error compensation model.

[0018] Further preferably, in S3, the azimuth angle and the pitch angle of the radar to the corner reflector are alternately transformed, and the millimeter wave radar angle error model is constructed, including:

[0019] Fix the pitch angle between the radar and the corner reflector to 0°, rotate the horizontal angle from -60° to 60°, with a step of 5°, for a total of 25 times, subtract the measured value obtained by the ranging error compensation model from the true value, and construct an optimal function model between the error and the horizontal angle;

[0020] Fix the horizontal angle to 0°, rotate the pitch angle from -30° to 30°, with a step of 5°, for a total of 13 times, subtract the measured value obtained by the ranging error compensation model from the true value, and construct an optimal function model between the error and the pitch angle.

[0021] Further preferably, the temperature and humidity near the radar are measured by the temperature and humidity meter, the relationship between the temperature, humidity and the compensated distance is analyzed, and the weather correction model is constructed, including the following steps:

[0022] Select any distance within 200 m from the radar, and measure the real-time weather factors including temperature and humidity values by the temperature and humidity meter and the timing photographing program;

[0023] The radar ranging value is corrected in sequence by the ranging error compensation model and the millimeter wave radar angle error model to obtain a compensated radar ranging distance;

[0024] The temperature value, humidity value and radar ranging distance are used to construct a mapping relationship matrix between the real-time weather and the corrected ranging value, forming a weather correction model.

[0025] Further preferably, in S5, the millimeter wave radar high-precision measurement model considering multiple influencing factors is constructed, including:

[0026] S501, set the approximate coordinates of any corner reflector, and obtain the millimeter wave distance observation value of the corner reflector according to Taylor series expansion;

[0027] S502, according to the observation equation when i radars simultaneously observe the same target at any time and the obtained millimeter wave distance observation value, the correction value of the unknown parameter is calculated by using nonlinear least squares estimation;

[0028] S503, according to the single observation radar pulse measurement precision difference, real distance and signal to noise ratio; comprehensive weighting is carried out, and a millimeter wave radar high precision measurement model considering various influence factors is established to settle the approximate coordinate estimation of the corner reflector.

[0029] Further preferably, in S501, the approximate coordinates of any corner reflector are set, and the millimeter wave distance observation value of the corner reflector is obtained according to Taylor series expansion.

[0030] The approximate coordinates of any corner reflector are set as The i-th radar coordinate is The first-order Taylor series expansion at has:

[0031]

[0032]

[0033] wherein, The millimeter wave distance observation value is represented; The millimeter wave radar reaches the geometric distance true value of the corner reflector; , , are intermediate variables; , , ; , , are coordinate increments of , , direction respectively.

[0034] Further preferably, in S502, according to the observation equation when the i-th radar simultaneously observes the same target at any time and the obtained millimeter wave distance observation value, the correction value of the unknown parameter is calculated by using nonlinear least square estimation, which includes:

[0035] The observation equation is ,

[0036] Let

[0037] ,

[0038] The observation equation is simplified as ;

[0039] The correction value of the unknown parameter is calculated by using nonlinear least square estimation:

[0040]

[0041]

[0042] In the formula, The initial values ​​of the parameters to be estimated are the approximate values ​​of the target coordinates. ; The obtained parameter correction number; Estimate the parameters obtained each time; It is a weight matrix.

[0043] More preferably, in S503, when performing comprehensive weighting based on the difference in measurement accuracy of a single observation radar pulse, the true distance, and the signal-to-noise ratio, the weighting formula is as follows:

[0044]

[0045] In the formula, Intermediate variables calculated to integrate the differences in radar pulse measurement accuracy, true distance, and signal-to-noise ratio in a single observation, i=1, 2, …, n, where n is the number of millimeter-wave radars. ; Let be the true distance between the i-th radar and target j in epoch k; The difference in pulse measurement accuracy between radar and target j in the i-th epoch of k epochs; Let be the signal-to-noise ratio of the i-th radar to target j in epoch k.

[0046] This application discloses a millimeter-wave radar measurement method that considers multiple influencing factors, effectively addressing the shortcomings of existing technologies. The invention utilizes the FFT frequency method and CZT to refine the ranging values ​​within the accuracy range. Addressing the insufficient ranging accuracy of millimeter-wave radar, it analyzes the correlation between distance and radar electronic component delay, distance and true distance, angle-related errors, and meteorological factors, constructing a refined ranging error compensation model to obtain accurate radar ranging values ​​for targets. Based on the compensated distance, azimuth, and elevation angles of the radar and corner reflectors, coarse target positioning is performed. A precise stochastic model is established by combining the differences in accuracy of single-observation radar pulse measurements, true distance, and signal-to-noise ratio. A spatial distance intersection method is used to measure the corner reflector position, achieving millimeter-level rapid measurement of track geometry within tunnels. This invention solves the problem of relying on foreign total station equipment for measurements in railway tunnels, improves the efficiency and accuracy of millimeter-wave radar measurements in complex tunnel environments, and effectively saves measurement costs. Attached Figure Description

[0047] Figure 1 This is a flowchart of a millimeter-wave radar measurement method that takes into account multiple influencing factors, provided by the present invention.

[0048] Figure 2 This is a flowchart of the millimeter-wave radar ranging refinement method in this invention. DETAILED DESCRIPTION

[0049] The application is further described in detail below with the aid of the accompanying drawings and specific embodiments.

[0050] As Figure 1 shown, the millimeter wave radar measurement method considering multiple influencing factors provided by an aspect of the application includes the following steps:

[0051] S1, obtaining original radar data and performing fine processing on the obtained original radar data to obtain an accurate ranging value of a single radar to a target; further preferably, as Figure 2 shown, in S1, the fine processing on the obtained original radar data to obtain an accurate ranging value of a single radar to a target includes:

[0052] S101, coarse extraction of a distance value, including: taking an angle reflector as a target, using FFT to calculate the frequency (mainly the wave peak with strong amplitude and concentrated frequency) of a discrete beat signal in original radar data, selecting the frequency corresponding to the signal with the maximum amplitude, and converting to obtain a target radial distance; the angle reflector is an eight-angle angle reflector. The calculation formula is as follows:

[0053]

[0054] In the formula, is the propagation speed of a signal in the atmosphere, is a sampling frequency, is a quantized frequency point after FFT, is a signal frequency modulation slope, is a complex sampling point.

[0055] The target radial distance (prior distance) obtained by calculation is used to determine a search interval, the signal with the maximum amplitude is selected, the point with the maximum amplitude in the search interval is determined, and the frequency fc corresponding to the maximum FFT amplitude is determined.

[0056] S102, spectral refinement, including coarse extraction of a target radial distance, determining the frequency corresponding to the target between the left and right adjacent spectral lines of the maximum spectral line (i.e., determining a spectral refinement interval according to the FFT result, selecting the maximum amplitude), and using CZT to refine the spectral line to obtain an unambiguous phase of the target (i.e., the frequency and phase corresponding to the maximum CZT amplitude); the start point of the refined frequency is the left spectral line adjacent to the maximum spectral line, and the end point is the right spectral line adjacent to the maximum spectral line, and the local spectrum is refined from the original 2 points to points. In order to avoid the problem of phase ambiguity, the phase difference between adjacent spectral lines should not exceed , and thus the refinement multiple is determined.

[0057] S103, range value refinement, according to the unambiguous phase of the target (i.e. the frequency and phase corresponding to the maximum CZT amplitude at the first epoch and the frequency and phase corresponding to the maximum CZT amplitude at the second epoch), the phase difference of the two range measurements is calculated, and the radial deformation distance is calculated according to the formula; the phase difference is converted into the corresponding distance change , which is superimposed into the first range value, to obtain the accurate range value of the next epoch.

[0058] The range value refinement is to calculate the phase difference of the two range measurements, and convert it into the corresponding distance change , which is superimposed into the first range value, to obtain the accurate range value of the next epoch.

[0059] The beat signal phase obtained by the radar measuring the corner reflector during the first range measurement is:

[0060]

[0061] The target deformation information is:

[0062]

[0063] In the formula, is the phase difference of the two range measurements, is the initial frequency of the signal, is the frequency modulation period of the signal, is the time delay of the signal from transmission to reception of the echo during the first range measurement, is the initial phase; is the quantized frequency point after FFT; therefore, the range error compensation model is represented by the following formula d t+1 =d t +△d .

[0064] S2, set the azimuth and elevation angles of the radar corner transmitter to 0; construct a range error compensation model;

[0065] wherein the azimuth and elevation angles of the radar corner transmitter are set to 0; a range error compensation model is constructed, including:

[0066] Adjust the elevation and horizontal angles of the radar antenna and the corner reflector to 0°, and move the corner reflector from a distance of 0.5m from the radar to 200m in a straight line with a step size of 0.1m, a total of 2000 times. The real absolute distance between the radar and the corner reflector is obtained by the Leica tracker AT960 cooperating with T-Pro as the true value, and the difference between the radar measured distance of the corner reflector and the true value is obtained as the optimal linear function model of the distance true value, as the range error compensation model.​​

[0067] S3, alternately transform the azimuth angle and the elevation angle of the corner reflector, and construct a millimeter wave radar angle error model;

[0068] Further, in S3, alternately transforming the azimuth angle and the elevation angle of the corner reflector, and constructing a millimeter wave radar angle error model comprises:

[0069] Fixing the elevation angle between the radar and the corner reflector as 0°, rotating the horizontal angle from -60° to 60°, with a step of 5°, a total of 25 times, and constructing an optimal function model between the error and the horizontal angle by subtracting the measured value obtained by using the ranging error compensation model from the true value.

[0070] According to the test data, the RMSE of the fitting result of the triangular function with the number of terms being 1 is the smallest, and the R-square is relatively high, so the optimal function model between the error and the horizontal angle is expressed by the following formula:

[0071]

[0072] x is the radar horizontal angle, and f(x) is the deviation of the radar ranging value from the true distance (after compensation of the ranging error model).

[0073] Fixing the horizontal angle as 0°, rotating the elevation angle from -30° to 30°, with a step of 5°, a total of 13 times, and constructing an optimal function model between the error and the elevation angle by subtracting the measured value obtained by using the ranging error compensation model from the true value.

[0074] Similarly, according to the test data, the RMSE of the fitting result of the first-order polynomial for the deviation between the true value and the measured value corresponding to the multiple collected elevation angles is the smallest, so the optimal function model between the error and the elevation angle is expressed by the following formula:

[0075]

[0076] x is the radar elevation angle, and f(x) is the deviation of the radar ranging value from the true distance (after compensation of the ranging error model).

[0077] S4, compensating the distance measured by the radar to the corner reflector by using the ranging error model and the angle error model, measuring the temperature and humidity near the radar by using a temperature and humidity meter, analyzing the relationship between the temperature, humidity and the compensated distance, and constructing a weather correction model.

[0078] Further preferably, the step of measuring the temperature and humidity near the radar by using a temperature and humidity meter, analyzing the relationship between the temperature, humidity and the compensated distance, and constructing a weather correction model comprises the following steps:

[0079] Select any distance within 200m from the radar, adopt the temperature and humidity meter and the timing photographing program to measure the real-time meteorological factors, the real-time meteorological factors including temperature value and humidity value;

[0080] The radar ranging value is corrected by the ranging error compensation model and the millimeter wave radar angle error model in sequence, and the compensated radar ranging distance is obtained;

[0081] The temperature value, humidity value and radar ranging distance are used to construct the mapping relationship matrix of real-time meteorological and corrected ranging value, and form the meteorological correction model.

[0082] According to the resolution of the 13th International Geodetic Conference, the calculation formula of the actual meteorological condition modulated carrier (light wave) refractive index n is:

[0083] (4-1)

[0084] In the formula, is the air expansion coefficient, and ; t is the actual atmospheric dry temperature, unit ℃; P is the actual atmospheric pressure, unit mmHg; e is the actual water vapor pressure, unit mmHg. n g is the refractive index of modulated light under standard meteorological conditions (T t =0℃, P =760mmHg, e =0mmHg), and its calculation formula is:

[0085] (4-2)

[0086] In the formula, Lambda is the monochromatic light wavelength, unit microns (μm); n λ is the refractive index of monochromatic non-modulated light under standard meteorological conditions, and its calculation formula is given by Barrell-Sears:

[0087] (4-3)

[0088] From the above formula, we can get:

[0089] (4-4)

[0090] From the above formula, we can get

[0091] (4-5)

[0092] For the millimeter wave radar used in this project, the wavelength Lambda =5000 μm, so the above formula can be used to calculate n g Substituting the refractive index calculation formula can be obtained:

[0093] (4-6)

[0094] In millimeter wave radar measurement, the air pressure is generally in millibar (mb), and the above formula is rewritten as:

[0095] (4-7)

[0096] The standard meteorological conditions of the millimeter wave radar in this project can be defined as: dry temperature t =12℃, atmospheric pressure P =1013.25mb, relative humidity h =60% (wet temperature is 8.3℃), and the reference refractive index n 0=1+276.5×10 -6 . Set:

[0097] (4-8)

[0098] (4-9)

[0099] The meteorological correction formula of the atmospheric refractive index on the distance measurement is:

[0100] (4-10)

[0101] In the formula, is the distance to be measured. At this point, the meteorological correction model related to the millimeter wave radar distance measurement can be established through the measured Euclidean distance, temperature, air pressure, humidity and other factors. The distance measurement value after the distance compensation, horizontal angle and pitch angle compensation correction is corrected using the meteorological correction model formula (2-40).

[0102] S5, according to the original radar data of S1, the initial coordinates are obtained, the distance error compensation model of S2, the millimeter wave radar angle error model of S3 and the meteorological correction model of S4 are used to obtain the corrected precise distance observation value, according to the influence of single pulse measurement error, distance length and signal-to-noise ratio on measurement, a millimeter wave radar high-precision measurement model considering multiple influencing factors is constructed, and the millimeter wave radar high-precision measurement data considering multiple influencing factors is obtained by using the millimeter wave radar high-precision measurement model.

[0103] Further preferably, in S5, the construction of the millimeter wave radar high-precision measurement model considering multiple influencing factors comprises:

[0104] S501, setting approximate coordinates of any corner reflector, according to Taylor series expansion, obtaining millimeter wave distance observation value of the corner reflector;

[0105] S502, according to the observation equation when i radar simultaneously observes the same target at any time and the obtained millimeter wave distance observation value, using nonlinear least square estimation, calculating the correction value of unknown parameter;

[0106] S503, according to the difference of single observation radar pulse measurement accuracy, real distance and signal to noise ratio; Comprehensive weighting is carried out, and a millimeter wave radar high-precision measurement model considering various influencing factors is established, and the approximate coordinate estimation of the corner reflector is settled.

[0107] Further preferably, in S501, setting approximate coordinates of any corner reflector, according to Taylor series expansion, obtaining millimeter wave distance observation value of the corner reflector includes:

[0108] Setting approximate coordinates of any corner reflector as , the i-th radar coordinate is , the first-order Taylor series expansion at has:

[0109]

[0110]

[0111] wherein, indicates the millimeter wave distance observation value; indicates the geometric distance value of the millimeter wave radar to the corner reflector; , , are intermediate variables; , , ; , , are coordinate increments of , , direction respectively.

[0112] Further preferably, in S502, according to the observation equation when i radar simultaneously observes the same target at any time and the obtained millimeter wave distance observation value, using nonlinear least square estimation, calculating the correction value of unknown parameter includes:

[0113] The observation equation is ,

[0114] Let

[0115] ,

[0116] The observation equation is abbreviated as ;

[0117] The correction value of the unknown parameter is calculated by using the nonlinear least square estimation:

[0118]

[0119]

[0120] In the formula, is the initial value of the set to be estimated parameter, that is, the approximate value of the target coordinate ; is the parameter correction number obtained; is the parameter estimation value obtained each time; is the weight matrix.

[0121] Further preferably, in S503, according to the single observation radar pulse measurement precision difference, the true distance and the signal-to-noise ratio; when the comprehensive weighting is performed, the weighting formula is as follows:

[0122]

[0123] In the formula, is an intermediate variable calculated by comprehensively considering the single observation radar pulse measurement precision difference, the true distance and the signal-to-noise ratio, i=1, 2, …, n, n is the number of millimeter wave radars, ; is the true distance of the kth epoch of the ith radar to the target j; is the pulse measurement precision difference of the kth epoch of the ith radar to the target j; is the signal-to-noise ratio of the kth epoch of the ith radar to the target j.

[0124] The precise distance observation value is obtained by correcting the radar distance measurement value by using the constructed distance measurement error model, the constructed angle error model and the constructed weather error model, and the approximate coordinate of the corner reflector is preliminarily estimated by combining the millimeter wave radar angle measurement method, and the calculation formula is as follows:

[0125]

[0126] In the formula, is the distance between the radar and the corner reflector, that is, the corrected radar distance measurement value; is the azimuth angle of the position of the corner reflector; is the pitch angle of the position of the corner reflector, that is, the radar directly measured and obtained.

[0127]

[0128] In the formula, Phase difference of echo signals of different receiving antennas For the distance between different receiving antennas.

[0129] The millimeter wave radar measurement method considering multiple influencing factors disclosed by the present application can effectively solve the deficiencies in the prior art. The present application uses FFT frequency method and CZT to refine the ranging values in the precision interval. In view of the problem of insufficient millimeter wave radar ranging accuracy, the correlation of distance with radar electronic element delay, with real distance, with angle-related error, and with meteorological factors is analyzed, a ranging error refinement compensation model is constructed, and accurate ranging values of the radar to the target are obtained. According to the compensated distance, azimuth angle and elevation angle of the radar and the corner reflector, the target is roughly positioned. A precise random model is established by combining the single observation radar pulse measurement accuracy difference, the real distance and the signal-to-noise ratio comprehensive weighting, and the spatial distance intersection method is used to measure the position of the corner reflector, so as to realize millimeter-level fast measurement of the track geometric state in the tunnel. The present application solves the problem that the measurement in the railway tunnel depends on foreign total station equipment, improves the millimeter wave radar measurement efficiency and accuracy in the complex tunnel environment, and effectively saves the measurement cost.

[0130] Obviously, the above embodiments are only examples for clearly illustrating, but not limit the embodiments. For ordinary skilled in the art, other different forms of changes or variations can be made on the basis of the above description. Here, all the embodiments need not and cannot be exhausted. The obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A method of millimeter wave radar measurement taking into account a plurality of influencing factors, characterized in that The method comprises the following steps: S1, obtaining original radar data and performing fine processing on the obtained original radar data to obtain an accurate ranging value of a single radar to a target; S2, setting the azimuth and elevation angles of the radar to the corner reflector to 0; constructing a ranging error compensation model; the ranging error compensation model construction process comprises: obtaining the real absolute distance between the radar and the corner reflector as a true value, and subtracting the distance measured by the radar from the corner reflector to obtain an optimal linear function model of the distance true value, which is the ranging error compensation model; S3, alternately transforming the azimuth and elevation angles of the radar to the corner reflector to construct a millimeter wave radar angle error model; the construction process of the millimeter wave radar angle error model comprises: fixing the elevation angle to 0°, adjusting the horizontal angle according to a preset horizontal angle start-stop range and step size, calculating the error between the measured value and the true value, and obtaining an optimal function model between the error and the horizontal angle; fixing the horizontal angle to 0°, adjusting the elevation angle according to a preset elevation angle start-stop range and step size, calculating the error between the measured value and the true value, and obtaining an optimal function model between the error and the elevation angle; S4, using the ranging error compensation model and the millimeter wave radar angle error model, compensating the distance obtained by the radar measuring the corner reflector, measuring the temperature and humidity near the radar by using a temperature and humidity meter, analyzing the relationship between the temperature, humidity and the compensated distance, and constructing a weather correction model; the weather correction model comprises: using the real-time measured temperature value, humidity value and compensated radar ranging distance to construct a mapping relationship matrix of real-time weather and corrected ranging value, forming a weather correction model; S5, obtaining an initial coordinate according to the original radar data of S1, obtaining a corrected accurate distance observation value by using the ranging error compensation model of S2, the millimeter wave radar angle error model of S3 and the weather correction model of S4, constructing a millimeter wave radar high-precision measurement model considering multiple influencing factors according to the influence of single pulse measurement error, distance length and signal-to-noise ratio on measurement, and obtaining millimeter wave radar high-precision measurement data considering multiple influencing factors by using the millimeter wave radar high-precision measurement model; comprising: S501, setting an approximate coordinate of any corner reflector, expanding according to Taylor series to obtain a millimeter wave distance observation value of the corner reflector; S502, using a nonlinear least squares estimation according to the observation equation when i radars simultaneously observe the same target at any time and the obtained millimeter wave distance observation value, calculating a correction value of an unknown parameter; S503, performing comprehensive weighting according to the single observation radar pulse measurement accuracy difference, real distance and signal-to-noise ratio, establishing a millimeter wave radar high-precision measurement model considering multiple influencing factors, and settling the approximate coordinate estimation of the corner reflector.

2. The method of claim 1, wherein, In S1, the fine processing on the obtained original radar data to obtain an accurate ranging value of a single radar to a target comprises: S101, distance value coarse extraction comprises: taking the corner reflector as a target, calculating the frequency of a discrete beat signal in the original radar data by using FFT, selecting the frequency corresponding to the maximum amplitude value, and converting to obtain the radial distance of the target; S102, spectrum refinement, including rough extraction of the target radial distance, determining the frequency corresponding to the target between the left and right adjacent spectrum lines of the maximum spectrum line, using CZT to refine the spectrum line to obtain the unambiguous phase of the target; S103, range value refinement, according to the unambiguous phase of the target, the phase difference of twice ranging is calculated, and the phase difference is converted into the corresponding distance change , the superimposed into the first ranging value, the next ephemeris accurate ranging value can be obtained. 3.The method of claim 1, wherein, In S2, the azimuth and elevation angles of the radar to the corner reflector are set to 0; a ranging error compensation model is constructed, including: Adjust the elevation and horizontal angles of the radar antenna and the corner reflector to 0°, and move the corner reflector from 0.5m to the radar at a step of 0.1m, a total of 2000 times, to 200m.

4. The method of claim 3, wherein the method further comprises: In S3, the azimuth and elevation angles of the radar to the corner reflector are alternately transformed, and a millimeter wave radar angle error model is constructed, including: Fix the elevation angle between the radar and the corner reflector to 0°, and rotate the horizontal angle from -60° to 60° at a step of 5°, a total of 25 times, and subtract the measured value obtained by the ranging error compensation model from the true value to construct an optimal function model between the error and the horizontal angle; Fix the horizontal angle to 0°, and rotate the elevation angle from -30° to 30° at a step of 5°, a total of 13 times, and subtract the measured value obtained by the ranging error compensation model from the true value to construct an optimal function model between the error and the elevation angle.

5. The method of claim 1, wherein, In S4, the temperature and humidity near the radar are measured by a temperature and humidity meter, the relationship between temperature, humidity and compensated distance is analyzed, and a weather correction model is constructed, including the following steps: Select any distance within 200m from the radar, and measure the real-time weather factors including temperature and humidity values by using a temperature and humidity meter and a timing photographing program; After the radar ranging value is corrected in turn by the ranging error compensation model and the millimeter wave radar angle error model, the compensated radar ranging distance is obtained; Using the temperature value, humidity value and radar ranging distance, a mapping relationship matrix between real-time weather and corrected ranging value is constructed to form a weather correction model.

6. The method of claim 1, wherein, In S501, set the approximate coordinates of any corner reflector, and according to Taylor series expansion, obtain the millimeter wave distance observation value of the corner reflector, including: Let the approximate coordinates of any corner reflector be set as , the i-th radar coordinate is , and at , the first-order Taylor series expansion has: wherein, denotes the millimeter wave range observation; denotes the millimeter wave angle of arrival reflector geometric range truth; , , are all intermediate variables; , , ; , , are the coordinate increments of the , , directions, respectively.

7. The method of claim 1, wherein the method further comprises: In S502, according to the observation equation when i radars simultaneously observe the same target at any time and the obtained millimeter wave distance observation value, the correction value of the unknown parameter is calculated by using nonlinear least squares estimation, including: The observation equation is , Let ， The observation equation is abbreviated as ; The correction value of the unknown parameter is calculated by using nonlinear least squares estimation: wherein is the initial value of the parameter to be estimated, i.e. the approximation of the target coordinates ; is the parameter correction obtained; is the parameter estimate obtained at each iteration; is the weight matrix.

8. The method of claim 1, wherein, In S503, according to the difference in single observation radar pulse measurement accuracy, the true distance and the signal-to-noise ratio; when the weights are comprehensively determined, the weight formula is as follows: wherein is an intermediate variable calculated to combine the difference in pulse measurement accuracy, the true range, and the signal-to-noise ratio of the i-th radar at the k-th epoch, i = 1, 2, …, n, n is the number of millimeter wave radars, ; is the true range of target j by the i-th radar at the k-th epoch; is the difference in pulse measurement accuracy of target j by the i-th radar at the k-th epoch; is the signal-to-noise ratio of target j by the i-th radar at the k-th epoch.

Images