Secondary radar deviation calibration method

Abstract

The invention discloses a secondary radar deviation calibration method. The method comprises the steps of coordinate transformation and target association, wherein to-be-calibrated radar track data and ADS-B track data are converted into a unified coordinate system with a radar center point as an original point, and target association is carried out; selecting a target track meeting a condition according to the radar track distribution; establishing a radar deviation model, and counting deviation of a target track to establish an error function; solving the deviation estimation of the radar through a cross iteration method; and smoothing the deviation value, and completing radar calibration through the smoothing deviation value. According to the method, the high-precision ADS-B track is taken as a reference, and the measurement difference between the secondary radar track and the reference track is gradually reduced by iteratively calculating the deviation, so that the correction effect is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of air traffic management, and particularly relates to a method for calibrating secondary radar deviation. Background Art

[0002] In the current air traffic management automation system, the fusion based on multiple sensors is crucial. Generally speaking, using multiple sensors in the automation system can improve the tracking performance. However, if there are some systematic deviations in the sensor measurements, the benefits of this multi-sensor fusion may be lost. The performance is manifested in that the sensor measurements with registration errors will reduce the tracking accuracy, and even cause false associations, resulting in splitting or false fusion, which will interfere with the controllers of the air traffic management automation system and easily trigger control safety incidents. Secondary radar is the most important sensor in the air traffic management automation system. Therefore, calibrating the secondary radar deviation is a prerequisite for multi-sensor fusion in the air traffic management automation system to ensure the accuracy of target tracking.

[0003] Automatic Dependent Surveillance-Broadcast technology, abbreviated as ADS-B, is a new generation of aviation target surveillance technology based on the Global Navigation Satellite System (GNSS). This technology captures GNSS signals through receivers on aircraft, accurately calculates the position information of the aircraft, and sends this position information in a broadcast form to ground ADS-B stations through on-board ADS-B transmitting equipment. After receiving this information, the ground stations further forward it to the air traffic control center, thereby realizing the effective surveillance of aircraft. Compared with traditional secondary radar, ADS-B has various advantages such as high-precision positioning ability, low operating cost, and high update frequency. Therefore, it has been widely used in China at present. According to the relevant technical specifications of civil aviation, the air traffic management automation system has clear requirements for the accessed ADS-B data. In particular, the quality factor is defined in the information transmission specification protocol CAT021 format, which can characterize the position accuracy level of the measurement data. When using high-precision ADS-B data to calibrate the deviation of secondary radar in real time, the measurement accuracy of a single radar for air targets will be greatly improved.

[0004] In traditional radar error research, the systematic errors in radar measurements mainly include ranging deviation and angle measurement deviation. The ranging deviation is a constant deviation in distance measurement, and the angle measurement deviation is the misalignment of the radar true north mark relative to the true north. It is usually assumed that these two deviations are invariant within the effective coverage area of the radar. However, in most cases, the convergence effects of these two deviations are not very satisfactory, and the convergence values show regional and time-varying characteristics. Summary of the Invention

[0005] Aiming at the deficiencies of the above-mentioned existing technologies, the purpose of the present invention is to provide a secondary radar deviation calibration method to solve the problem of unstable convergence value of calibration error in the existing technologies. The method of the present invention takes the high-precision ADS-B track as the benchmark, and gradually reduces the measurement difference between the secondary radar track and the benchmark track through iterative calculation of the deviation, so as to achieve the calibration effect.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0007] A secondary radar deviation calibration method of the present invention comprises the following steps:

[0008] 1) Coordinate transformation and target association: Convert the radar track data to be calibrated and the ADS-B track data into a unified coordinate system with the radar center point as the origin and perform target association;

[0009] 2) Select target tracks that meet the conditions according to the radar track distribution;

[0010] 3) Establish a radar deviation model, and count the deviation of the target tracks to establish an error function;

[0011] 4) Solve the deviation estimation of the radar by the cross-iteration method;

[0012] 5) Smooth the deviation value and complete the radar calibration through it.

[0013] Further, in the step 1), an association criterion is established by the track recognition information elements (address code, secondary code, flight number, etc.) and spatial position elements (horizontal position, altitude, course, etc.) of the target, and the ADS-B track is associated with the radar track.

[0014] Further, the association criterion in the step 1) is specifically as follows: Compare the consistency of each element of the two tracks. The consistency of the recognition information elements means that the recognition elements are completely the same, and the consistency of the spatial position elements means that the difference of the spatial elements meets a certain range constraint, specifically, the horizontal position distance difference is less than a distance threshold, the altitude difference is less than an altitude threshold, and the course difference is less than an angle threshold; According to the processing logic sequence of first the recognition information elements and then the spatial position elements, when a comparison element is missing in the track, it is default to pass and enter the comparison of the next element until all elements are compared; When there is an inconsistency in a certain element, the association fails, and when there is no inconsistent element, the association succeeds.

[0015] Further, the step 2) specifically includes: Check the position distribution of the radar tracks associated with the ADS-B tracks in the step 1), limit their positions to be simultaneously within the effective coverage ranges of the radar site and the ADS-B site, and select target tracks that maintain uniform linear motion in the symmetric quadrants of the radar center.

[0016] Further, step 3) specifically includes: sorting out the data of the target track selected in step 2), filtering out the position point data with abnormal swings, performing time compensation on the ADS-B track signals associated with the same target, aligning the time to the radar signal update moment, and simultaneously checking the position accuracy index NACp (Navigational Accuracy Category for Position) of the ADS-B track signal at this moment. When the position navigation accuracy category is higher than / equal to 10, the horizontal position element in the ADS-B track can be used as the true position of the target; in the radar center rectangular coordinate system, use (x adsb , y adsb ) to represent the position of the ADS-B track after time alignment processing, use (x rad , y rad ) to represent the position of the radar track associated with the same target. The position difference between the two tracks at the same moment is decomposed into the radial deviation D ρ in the radial direction of the polar radius ρ and the axial deviation D θ in the direction of the polar angle θ in the radar polar coordinate system (ρ, θ). Use b ρ , b θ to represent the ranging deviation and angle measurement deviation of the radar respectively, and use b t to represent the time deviation caused by data transmission and processing. Establish the deviation models of the radial deviation D ρ and the axial deviation D θ with respect to b ρ , b θ and b t as follows:

[0017]

[0018] Among them, ω ρ and ω θ are respectively the random error independent in the radial direction and the random error independent in the axial direction. When calculating the axial deviation, due to the product effect of the angle measurement deviation and the radial distance, the calculation accuracy of the angle measurement deviation will be reduced when accumulating the target deviations at different radial distances. Normalize the axial deviation with respect to the preset polar radius distance ρ c . The transformation result is:

[0019]

[0020] According to the plane xy coordinates of the radar track and the ADS-B track under a single trace of the same target, solve the radial deviation D ρ and the axial deviation D θ . The expressions are as follows:

[0021]

[0022] Substitute it into the deviation model and construct an error function from the random error as follows:

[0023]

[0024] Among them, J represents the random error function when updating a single point of a single target;

[0025] Model all moving points of the target within a sliding interval to update the error function, expressed as Model and update the error function for all the counted targets (total number N) to solve the final b ρ , b θ and b t such that the overall error function is minimized, where M is the number of points in the sliding interval, N is the total number of counted targets, E is the overall error function, i is the target serial number, and j is the point serial number.

[0026] Furthermore, step 4) specifically includes:

[0027] Set the ranging deviation b ρ and the angle measurement deviation b θ as constants, and use the first-order partial derivative of the error function E with respect to the time deviation b t equal to 0 to solve the extreme point of the error function E, corresponding to the initial value of the time deviation estimate as follows:

[0028]

[0029] After deducting the influence of the initial value of the time deviation estimate ρ from the radial deviation D θ and the axial deviation D , obtain the remaining radial deviation D ρ1 and the remaining axial deviation D θ1 , and update the j-th point of the i-th target of the radar. The expression is as follows:

[0030]

[0031] Set the remaining time deviation as b t2 , and after removing the influence of the initial value of the time deviation estimate , update the error function:

[0032]

[0033] Take the partial derivatives of b ρ and b θ , and set the partial derivatives equal to 0 to find the extreme points, corresponding to the initial value of the ranging deviation estimate and the initial value of the angle measurement deviation estimate as follows:

[0034]

[0035] Remove the initial ranging deviation estimate value ρ1 and the initial azimuth deviation estimate value ρ1 from the remaining radial deviation D and the remaining azimuth deviation D to update and obtain new remaining deviations to replace D ρ and D θ , repeat the cross-iteration process to obtain the remaining time deviation estimate value of the second iteration remaining ranging deviation estimate value and remaining azimuth deviation estimate value Continue the iterative calculation to obtain a new remaining deviation estimate, which forms a geometrically decreasing sequence, and the decreasing factor K satisfies the following:

[0036]

[0037] Sum up the initial deviation estimate value calculated in the first iteration and the remaining deviation estimate values of each subsequent iteration to obtain the total deviation estimate value.

[0038] Furthermore, the step 5) specifically includes: using a first-order lag smoothing mechanism to reduce the influence of random noise of different statistical targets under different measurement update cycles, and the first-order lag smoothing mechanism is as follows:

[0039]

[0040] where α is the smoothing coefficient; is the deviation estimate value under the measurement update cycle k, and the subscript cor represents any one of the time deviation, ranging deviation, and azimuth deviation; b s (k), b s (k - 1) are the deviation smoothing values of the corresponding time deviation, ranging deviation, or azimuth deviation under the measurement cycle k and the measurement cycle k - 1 respectively; use b s (k) to perform radar calibration processing to obtain a new measurement value, complete the radar calibration, and for the i-th radar target, the calibration form is as follows:

[0041]

[0042] where b ts , b ρs and b θs represent the smoothing values of the current measurement cycle time deviation, ranging deviation, and azimuth deviation respectively, and represent the new polar radius measurement and polar angle measurement obtained after calibration processing respectively, and T rad represents the radar update cycle.

[0043] Advantages of the present invention:

[0044] The present invention represents the true position and speed of the target through high-precision ADS-B data, establishes a deviation model in the local coordinate system of the radar to be calibrated, decomposes the radar measurement deviation into range measurement deviation, angle measurement deviation and time deviation, estimates the deviation through the cross-iteration solution method within the sliding interval, combines the smoothing algorithm to smooth the estimated value, and uses the smoothed deviation value to compensate the radar to be calibrated, thereby completing the calibration of the radar to be calibrated.

[0045] The method of the present invention combines high-precision ADS-B signals to select appropriate radar tracks, improves the stability of radar deviation convergence by introducing time deviation, the solution equation and calculation steps are simple, and it can be applied in real time, which is beneficial to improving the tracking accuracy of the integrated track in the air traffic management automation system. Brief Description of the Drawings

[0046] Figure 1 is a flowchart of the method of the present invention.

[0047] Figure 2 is a schematic diagram of the target selection distribution of the calibration radar.

[0048] Figure 3 is a schematic diagram of the deviation model between the radar track and the ADS-B track under a single target plot.

[0049] Figure 4 is a schematic diagram of the target trajectories of the radar track and the ADS-B track in the implementation case.

[0050] Figure 5 is a schematic diagram of the radar deviation estimation result in the implementation case.

[0051] Figure 6 is a schematic diagram of the comparison result of the average deviation statistics before and after the radar deviation correction in the implementation case. Detailed Embodiment

[0052] For the convenience of understanding by those skilled in the art, the present invention will be further described below in conjunction with the embodiments and the drawings. The content mentioned in the embodiments does not limit the present invention.

[0053] Referring to Figure 1 as shown, a method for calibrating the deviation of a secondary radar according to the present invention is as follows:

[0054] 1) Coordinate transformation and target association: Convert the radar track data to be calibrated and the ADS-B track data into a unified coordinate system with the radar center point as the origin and perform target association;

[0055] The radar track measurement data is converted from polar coordinates to planar coordinates, and the ADS-B track measurement data is transformed from longitude and latitude to planar coordinates through Gauss projection transformation; after the conversion, the correlation relationship between the radar track and the ADS-B track is established. The said correlation relationship is whether they belong to the same target. The correlation criterion is established based on the track identification information elements (address code, secondary code, flight number, etc.) and spatial position elements (horizontal position, altitude, heading, etc.) of the target, and the ADS-B track and the radar track are target-correlated.

[0056] 2) When calculating the deviation, it is necessary to count the targets in different regions to avoid the deviation estimation from falling into local convergence in a single region. At the same time, in order to reduce the influence of the asymmetry error and the radar (low altitude and top altitude) blind area, as well as the measurement difference caused by the speed change, the selected targets need to meet the conditions that they are within the effective coverage area jointly covered by the radar and ADS-B and the track of the target is a uniform linear motion that conforms to the symmetric quadrants of the radar coordinate system; the effective radar target distribution is as Figure 2 shown, located in regions A1, B1, A2, and B2 respectively. The targets in A1 and B1, and A2 and B2 are paired to form the selected targets for deviation calculation.

[0057] 3) Establish a radar deviation model, and count the deviation of the target track to establish an error function;

[0058] Sort out the data of the target track selected in step 2), filter out the position point data with abnormal swings, perform time compensation on the ADS-B track signal associated with the same target, align the time to the radar signal update moment, and at the same time check the position accuracy index NACp (Navigational Accuracy Category for Position) of the ADS-B track signal at this moment. When the position navigation accuracy category is higher than / equal to 10, the horizontal position element in the ADS-B track can be used as the true position of the target; in the radar center rectangular coordinate system, use (x adsb , y adsb ) to represent the position of the ADS-B track after time alignment processing, use (x rad , y rad ) to represent the position of the radar track associated with the same target. The position difference between the two tracks at the same moment is decomposed into the radial deviation D ρ in the radial direction of the polar radius ρ and the axial deviation D θ in the direction of the polar angle θ in the radar polar coordinate system (ρ, θ). Use b ρ , b θ to represent the ranging deviation and the angle measurement deviation of the radar respectively, and use b t to represent the time deviation caused by data transmission and processing. Establish the radial deviation D ρ , the axial deviation D θ with respect to bρ , b θ and b t The deviation models of are as follows:

[0059]

[0060] where ω ρ and ω θ are respectively the radially independent random error and the axially independent random error (when calculating the axial deviation, due to the product effect of the angular measurement deviation and the radial distance, the calculation accuracy of the angular measurement deviation will be reduced when accumulating the target deviations at different radial distances). The pre-specified polar radius distance ρ c is normalized for the axial deviation, and the transformation result is:

[0061]

[0062] According to the planar xy coordinates of the radar track and the ADS - B track under a single point trace of the same target, the radial deviation D ρ and the axial deviation D θ are solved, and the expressions are as follows:

[0063]

[0064] Substituting into the deviation model, an error function is constructed from the random error, as follows:

[0065]

[0066] where J represents the random error function during the update of a single point trace of a single target;

[0067] Modeling all the moving point traces of the target within a sliding interval to update the error function, expressed as Modeling and updating the error function for all the statistically counted targets (total number N) to solve the final b ρ , b θ and b t such that the overall error function is minimized, where M is the number of point traces in the sliding interval, N is the total number of statistically counted targets, E is the overall error function, i is the target serial number, and j is the point trace serial number.

[0068] 4) Solve the deviation estimation of the radar through the cross - iteration method;

[0069] Set the ranging deviation b ρ and the angular measurement deviation b θ as constants, and use the first - order partial derivative of the error function E with respect to the time deviation b t equal to 0 to solve the extreme point of the error function E, corresponding to the initial value of the time deviation estimation as follows:

[0070]

[0071] After deducting the influence of the initial time deviation estimate from the radial deviation D ρ and the axial deviation D θ , the remaining radial deviation D and the remaining axial deviation D ρ1 are obtained. For the j-th trace update of the i-th target of the radar, the expression is as follows: θ1 Set the remaining time deviation as b

[0072]

[0073] , and after removing the influence of the initial time deviation estimate t2 , update the error function:

[0074]

[0075] Take the partial derivatives of b ρ and b θ , and set the partial derivatives equal to 0 to find the extreme points, corresponding to the initial ranging deviation estimate and the initial angle measurement deviation estimate as follows:

[0076]

[0077] In the remaining radial deviation D ρ1 and the remaining axial deviation D ρ1 , remove the influence of the initial ranging deviation estimate and the initial angle measurement deviation estimate . Update to obtain the new remaining deviation to replace D ρ and D θ . Repeat the cross-iteration process to obtain the remaining time deviation estimate value of the second iteration , the remaining ranging deviation estimate value and the remaining angle measurement deviation estimate value . Continue the iterative calculation to obtain a new remaining deviation estimate, which forms a geometrically decreasing sequence, and the decreasing factor K satisfies the following:

[0078]

[0079] Sum up the initial deviation estimate of the first iterative calculation and the remaining deviation estimates of each subsequent iteration to obtain the total deviation estimate.

[0080] 5) Smooth the deviation value and complete the radar calibration through it.

[0081] ​Reduce the influence of random noise of different statistical targets under different measurement update periods by using a first-order lag smoothing mechanism. The first-order lag smoothing mechanism is as follows:

[0082]

[0083] where α is the smoothing coefficient; is the deviation estimation value at the measurement update period k, and the subscript cor represents any one of the time deviation, ranging deviation, and angle measurement deviation; b s (k), b s (k - 1) are the deviation smoothing values of the corresponding time deviation, ranging deviation, or angle measurement deviation at the measurement period k and the measurement period k - 1 respectively; Use b s (k) to perform radar calibration processing to obtain a new measurement value, complete radar calibration. For the i-th radar target, the calibration form is as follows:

[0084]

[0085] where b ts , b ρs , and b θs represent the smoothing values of the current measurement period time deviation, ranging deviation, and angle measurement deviation respectively, and represent the new polar radius measurement and polar angle measurement obtained after calibration processing respectively, and T rad represents the radar update period.

[0086] Through the above steps, dynamic calibration of the deviation of the radar to be calibrated can be achieved. Due to the small number of statistical targets (usually only a dozen or so), the summation operation amount is small and the number of iteration steps is small, so it has strong real-time performance. The radar measurement value after deviation calibration can be applied to the fusion processing of the subsequent automation system in time, improving the tracking accuracy of the system fusion track.

[0087] For further illustration by example, consider an automated system that simulates a secondary radar and an ADS-B track. The radar center is located at the origin, and the running trajectory points of some targets are as shown in Figure 3 . It can be clearly seen from the figure that there are certain differences between the two signal trajectories of the target, and the differences are more obvious far from the origin. By solving the deviation through the above steps, the analysis results are as shown in Figure 4 . Since there is no obvious lead-lag phenomenon between the two tracks, the calculated time deviation estimation is very small and not shown in the figure; the ranging deviation estimation and the angle measurement deviation estimation fluctuate slightly within a certain range. According to the results of Figure 4 , perform radar track deviation correction to obtain a new radar measurement value, and statistically calculate the average deviation of a certain number of targets before and after correction. Referring to Figure 6 shown, the average deviation calculation method is as follows:

[0088]

[0089] The comparison of the err differences before and after deviation correction is as Figure 5 shown. It can be seen that after correction, err decreases significantly, from 1.4 km to 0.2 km, a decrease of about 86%. Therefore, the tracking accuracy of the integrated track can be significantly improved after correction.

[0090] There are many specific application ways of the present invention. The above description is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements can be made, and these improvements should also be regarded as the protection scope of the present invention.

Claims

1. A secondary radar deviation calibration method, characterized in that: Here are the steps: 1) Coordinate transformation and target association: transform the radar track data to be calibrated and the ADS-B track data into a unified coordinate system with the radar center point as the origin and perform target association; 2) Select target tracks that meet the requirements based on radar track distribution; 3) Establish a radar deviation model and calculate the deviation of the target track to establish an error function; 4) Solve the radar deviation estimation by cross iteration method; 5) Smooth the deviation value and use it to complete the radar calibration.

2. The secondary radar deviation calibration method according to claim 1, characterized in that: In the step 1), an association criterion is established based on the target's track identification information elements and spatial position elements, and the ADS-B track and the radar track are target associated.

3. The secondary radar deviation calibration method according to claim 2, characterized in that: The association criteria in step 1) are specifically as follows: comparing the consistency of the elements of the two tracks, wherein the consistency of the identification information elements refers to the complete consistency of the identification elements, and the consistency of the spatial position elements refers to the difference in the spatial elements satisfying a certain range constraint, specifically, the horizontal position distance difference is less than a distance threshold, the height difference is less than a height threshold, and the heading difference is less than an angle threshold; according to the processing logic order of first identifying the information elements and then the spatial position elements, when a comparison element is missing in the track, the comparison of the next element is entered by default until all elements are compared; when an element is inconsistent, the association fails, and when there is no inconsistent element, the association succeeds.

4. The secondary radar deviation calibration method according to claim 1, characterized in that: The step 2) specifically includes: checking the position distribution of the radar track successfully associated with the ADS-B track in step 1), limiting its position to be within the effective coverage range of the radar site and the ADS-B site at the same time, and selecting the target track that maintains uniform linear motion in the symmetrical quadrant of the radar center.

5. The secondary radar deviation calibration method according to claim 1, characterized in that: The step 3) specifically includes: sorting the target track selected in step 2), filtering out the abnormal swinging position point data, performing time compensation on the ADS-B track signal associated with the same target, aligning the time to the radar signal update time, and checking the position accuracy index NACp of the ADS-B track signal at this time. When the position navigation accuracy category is higher than / equal to 10, the horizontal position element in the ADS-B track can be used as the real position of the target; in the radar center rectangular coordinate system, (x adsb ,y adsb ) represents the ADS-B track position after time alignment, and (x rad ,y rad ) represents the position of the radar track associated with the same target. The position difference between the two tracks at the same time is decomposed into the radial deviation D in the direction of the polar radius ρ in the radar polar coordinate system (ρ, θ) ρ Axial deviation D in the direction of polar angle θ θ , use b ρ 、b θ They represent the ranging deviation and angle measurement deviation of the radar respectively, and b t Indicates the time deviation caused by data transmission and processing, and establishes the radial deviation D ρ , axial deviation D θ About b ρ 、b θ and b t The deviation model is as follows: Among them, ω ρ and ω θ They are radial and axial independent random errors, respectively, and the axial deviation is pre-specified by the polar distance ρ c Normalized, the transformation result is: The radial deviation D is calculated based on the plane xy coordinates of the radar track and ADS-B track under a single point track of the same target. ρ and axial deviation D θ , the expression is as follows: Bring it into the deviation model and construct the error function from the random error as follows: Where J represents the random error function when updating a single point of a single target; Model all the moving points of the target in a sliding interval to update the error function, which is expressed as Model the statistical targets and update the error function to solve the final b ρ 、b θ and b t , so that the overall error function To achieve minimization, where M is the number of points in the sliding interval, N is the total number of targets counted, E is the overall error function, i is the target number, and j is the point number.

6. The secondary radar deviation calibration method according to claim 5, characterized in that: The step 4) specifically includes: Set the distance measurement deviation b ρ and angle measurement deviation b θ is a constant, and the error function E is used to calculate the time deviation b t The first-order partial derivative of is equal to 0, and the extreme point of the error function E is solved, corresponding to the initial value of the time deviation estimate as follows: The radial deviation D ρ and axial deviation D θ Deducting time deviation to estimate the initial value After the influence of ρ1 and residual axial deviation D θ1 , update the jth point trace of the i-th target of the radar, the expression is as follows: Set the remaining time deviation to b t2 , remove the time deviation to estimate the initial value After the influence of , update the error function: For b ρ and b θ Find the partial derivative and set it equal to 0, find the extreme point, and estimate the initial value of the corresponding ranging deviation and the initial value of the angle measurement deviation as follows: The residual radial deviation D ρ1 and residual axial deviation D ρ1 Remove the ranging bias to estimate the initial value and the initial value of the angle measurement deviation The influence of the update is to get a new residual deviation to replace D ρ and D θ , repeat the cross-iteration process to obtain the remaining time deviation estimate of the second iteration Remaining ranging bias estimate and the residual angular deviation estimate Continue iterating to obtain a new residual deviation estimate, which is a geometrically decreasing sequence, and the decreasing factor K satisfies the following: The total deviation estimate is obtained by summing up the initial deviation estimate calculated in the first iteration and the residual deviation estimate of each subsequent iteration.

7. The secondary radar deviation calibration method according to claim 1, characterized in that: The step 5) specifically includes: using a first-order hysteresis smoothing mechanism to reduce the influence of random noise of different statistical targets under different measurement update cycles, and the first-order hysteresis smoothing mechanism is as follows: Among them, α is the smoothing coefficient; is the estimated value of the deviation under the measurement update period k, and the subscript cor represents any one of the time deviation, ranging deviation and angle measurement deviation; b s (k), b s (k-1) are the deviation smoothing values ​​of the corresponding time deviation, distance deviation or angle deviation under the measurement period k and measurement period k-1 respectively; using b s (k) Perform radar calibration processing to obtain new measurement values ​​and complete radar calibration. For the i-th radar target, the calibration form is as follows: Among them, b ts 、b ρs and b θs They represent the smoothed values ​​of the current measurement cycle time deviation, distance measurement deviation and angle measurement deviation, respectively. and They represent the new polar diameter measurement and polar angle measurement after calibration, respectively. rad Indicates the radar update period.

Images