A method for joint calibration of azimuth-elevation errors of a distributed radar-optoelectronic system

Abstract

The application provides a distributed radar photoelectric system azimuth-elevation error joint calibration method, which is suitable for the anti-unmanned aerial vehicle field.The application divides the azimuth space of the plane where the radar photoelectric system is located into six quadrants, and calibrates them one by one, releases the unmanned aerial vehicle at the middle position of each azimuth quadrant, counts the detection azimuth data and elevation data of the radar and the photoelectric system, calculates the azimuth difference and the elevation difference between them, divides the data into several distance sections according to the fluctuation of the azimuth difference curve and the elevation difference curve, calculates the azimuth difference average and the elevation difference average of each distance section, and then calibrates the azimuth and the elevation of each distance section one by one.The azimuth difference and the elevation difference of the whole data after calibration are lower than one third of the azimuth visual angle of the photoelectric system, and are judged as qualified, otherwise, the calibration needs to be re-performed.

Description

Technical Field

[0001] This invention is applicable to the field of anti-drone technology. Background Technology

[0002] Radar and optoelectronic devices each have their own characteristics. Radar has strong capabilities in airspace surveillance and target detection, while optoelectronic devices offer high target tracking accuracy, high data rates, and the ability to acquire target image information. Radar often guides optoelectronic devices for tracking and imaging after detecting a target. For distributed radar-optoelectronic equipment, there is often a certain installation gap between the radar and optoelectronic devices. The installation error between the radar and optoelectronic devices is amplified within the detection space due to this gap, and the deviation is not constant within the space. Without proper handling, situations may arise where the radar cannot guide the optoelectronic device into the field of view. Furthermore, because optoelectronic devices have high measurement accuracy, if the horizontal alignment of the mounting bases of the two devices is insufficient, even without considering the installation gap, situations may arise where the radar cannot successfully guide the optoelectronic device to track the target into the field of view. Additionally, situations may occur where the radar can guide the optoelectronic device into the field of view in a specific azimuth, while radars in other directions cannot. This is caused by inconsistent horizontal alignment of the mounting bases, resulting in the radar and optoelectronic devices not being aligned within the 360-degree azimuth range. Considering both of these factors, for distributed radar-electro-optical systems, where the horizontal consistency of the base cannot be guaranteed to be below 0.1°, it is necessary to find a way to correct the detection data of the radar and the electro-optical system in order to ensure that the radar guides the electro-optical system to enter the field of view smoothly.

[0003] Current research primarily focuses on ideal scenarios for radar and electro-optical calibration. The patent "Target Elevation Angle Acquisition Method and Device Based on Radar-Electro-Optical Linkage System" addresses the elevation angle alignment problem of radar and electro-optical linkage devices. It mainly focuses on the general principles, methods, and processes for acquiring target elevation data from radar and electro-optical linkage devices, which is relevant for data correction of individual azimuth angles. However, it does not address how to ensure radar and electro-optical data alignment at every angle within a 360° azimuth range when the mounting base is not level. Summary of the Invention

[0004] This invention proposes a joint calibration method for azimuth and elevation errors of a distributed radar-electro-optical system to solve the problem of aligning radar and electro-optical data and guiding the radar to smoothly enter the field of view in a 360° omnidirectional range when the radar and electro-optical base cannot be adjusted to be completely horizontal.

[0005] To achieve the above objectives, the steps of the present invention are as follows:

[0006] Step 1: Divide the azimuth space where the radar and optoelectronic equipment are located into 6 quadrants;

[0007] Step 2: Calibrate the data quadrant by quadrant. First, select the first quadrant for calibration, including:

[0008] 1) Launch the UAV at the 30° azimuth of the middle position of the azimuth interval. The radar and electro-optical systems will respectively detect and obtain the target's azimuth and elevation angle values. The target's flight distance should be from 100 meters beyond the radar blind zone to 80% of the maximum detection power of the electro-optical system. Record two sets of data: one for the position far from the equipment and one for the position close to the equipment.

[0009] 2) Select the stationary and continuous data set from the two sets for use. First, remove outliers, then perform error statistics. The statistical method is as follows:

[0010] The radar track and electro-optical track data are recorded according to the radar's rotation speed. Assume the radar rotation speed is ts, the number of recorded data points is N, and the azimuth acquired by the radar is θ. Rn The elevation angle is β Rn The azimuth obtained by photoelectric detection is θ Pn The elevation angle is β Pn The azimuth difference is denoted as △θ n The difference in elevation angle is denoted as △β. n , n∈[1,N];

[0011] △θ n =θ Rn -θ Pn

[0012] △β n =β Rn -β Pn

[0013] R∈[ts,ts*N]

[0014] 3) Plot the distance-azimuth difference (R, Δθ) respectively. n Curve and distance-elevation difference (R, Δβ) n )curve;

[0015] 4) Mark data points that differ from surrounding data by more than 0.5 as outliers, change the outlier value to be the same as the previous data point, and redraw the distance-azimuth difference (R, Δθ). n Curve and distance-elevation difference (R, Δβ) n )curve;

[0016] 5) Segment, statistically analyze, and correct the positional data. The specific methods are as follows:

[0017] First observe the distance-azimuth difference (R, Δθ) nThe curve is used to segment the data according to its fluctuation trend. A total of N data points are divided into K segments, denoted as k, where k∈[1,K]. The number of data points in each segment is denoted as N. k The data sequence number within a segment is denoted as nk, and the azimuth difference data for each segment is denoted as Δθ. nk The mean azimuth difference of the radar electro-optical signal within each segment is calculated as follows: k∈[1,K];

[0018] Assuming the normal azimuth detection data of photoelectric sensors at each distance segment within this quadrant are: The photoelectric orientation is corrected for each distance segment, and the corrected data is recorded as follows:

[0019] 6) The elevation angle difference data is segmented, statistically analyzed, and corrected. The specific methods are as follows:

[0020] In the observation distance-elevation angle difference (R, Δβ) n The curve is used to segment the data according to its fluctuation trend. A total of N data points are divided into K segments, denoted as k, where k∈[1,K]. The number of data points in each segment is denoted as N. k The data sequence number within a segment is denoted as nk, and the difference in elevation angle for each segment is denoted as Δβ. nk The mean elevation angle difference of the radar electro-optical system within each segment is calculated as follows: k∈[1,K];

[0021] Assuming the normal elevation angle detection data of photoelectric sensors for each distance segment within this quadrant are: The elevation angle of the photoelectric sensor is corrected for each distance segment, and the corrected data is recorded as follows:

[0022] Step 3: Fly another flight, and follow steps 1) to 3) in Step 2 to calculate the azimuth and elevation errors between the radar and electro-optical systems, and calculate the mean azimuth and elevation errors. If the mean azimuth error is less than one-third of the electro-optical azimuth field of view, and the mean elevation error is less than one-third of the electro-optical azimuth field of view, the calibration for the current quadrant ends, proceed to Step 4; otherwise, proceed to Step 2.

[0023] Step 4: Launch the drone at the midpoint of each quadrant from 2 to 6, and repeat steps 2 and 3 to complete the data measurement, calculation and correction for the current quadrant.

[0024] Furthermore, in the first step, the line connecting the installation positions of the radar and optoelectronic equipment is used as a reference. The entire airspace is divided into six quadrants on average for data measurement, and each quadrant is corrected according to the same set of correction values. The end where the optoelectronic and radar rays are located is designated as the azimuth 0, the end where the radar and optoelectronic rays are located is designated as the azimuth 3, and the starting point is the line where the radar is located. The radar is marked 60° clockwise from its starting point as the azimuth 1. The starting point clockwise 120° is marked as bearing 2, the starting point clockwise 240° is marked as bearing 4, and the starting point clockwise 300° is marked as bearing 5; these form 6 quadrants from 1 to 6: Quadrant 1: 0-60°, Quadrant 2: 60°-120°, Quadrant 3: 120°-180°, Quadrant 4: 180°-240°, Quadrant 5: 240°-300°, and Quadrant 6: 300°-360°.

[0025] Furthermore, in the second step, the more stable and continuous set of data is selected from the two sets of data, one near the device and one far from the device. The selection method is to select data with a radar track loss rate of less than 10% and an electro-optical track loss rate of less than 10%. If neither can be met, the UAV needs to be re-launched to acquire data. The track loss rate is calculated as follows: the number of data points that should be in the flight distance segment is denoted as N_all, the number of track data points actually recorded is N_Track, and the track loss rate Track-Loss-Rate is calculated as: Track-Loss-Rate=(N_all-N_Track) / N_all.

[0026] Furthermore, in the second step, the difference is calculated by statistical data segmenting the distance according to the distribution of the distance-azimuth difference curve and the distance-elevation difference. Data with a azimuth difference or elevation difference fluctuation exceeding 0.3 degrees are divided into segments.

[0027] This invention addresses the challenge of calibrating radar and photoelectric linkage systems with a certain distance between their installation locations, where the horizontal consistency of the base cannot be guaranteed to be below 0.1°. It proposes a specific engineering-based calibration method to ensure that the radar guides the photoelectric system smoothly into the field of view. This invention successfully solves the calibration problem for separately arranged radar and photoelectric systems under conditions of limited flatness, demonstrating its engineering application value. Attached Figure Description

[0028] Appendix Figure 1 Flowchart of azimuth and elevation calibration for a distributed radar-electro-optical linkage system;

[0029] Appendix Figure 2 Orientation sector division map;

[0030] Appendix Figure 3 Schematic diagram showing the location of the directional target;

[0031] Appendix Figure 4 Assume a distance-azimuth difference curve;

[0032] Appendix Figure 5 Assume a distance-elevation angle difference curve. Detailed Implementation

[0033] The invention will be further explained below with reference to the accompanying drawings. The preferred implementation process of the distributed radar electro-optical system azimuth and elevation error joint calibration method proposed in this invention is shown in the attached figure. Figure 1 As shown, the specific description is as follows:

[0034] As attached Figure 2 As shown, the installation locations of the radar and photoelectric equipment are connected by a straight line. The center of the two devices is taken as the origin, and the line connecting the origin to the radar is taken as the azimuth 0. The azimuth space is divided into six 60° sectors in a counterclockwise direction.

[0035] Assuming the radar's short-range blind zone is 150m, the electro-optical detection range is 3000m, and the UAV's flight range is 250m to 2400m from the radar, with the flight direction as shown in the attached figure. Figure 3 As shown by the dotted lines, the center of each sector is where the aircraft flies first in direction a (from the inside out), then in direction b (from the outside in), collecting radar and electro-optical tracking data.

[0036] Statistically analyze the flight data of two groups, A and B. Assuming the radar cycle is 2 seconds, record radar tracks and electro-optical tracks according to the radar data rate. Calculate the UAV flight speed as 10 m / s. Then, there should be 107 radar tracks and 107 electro-optical tracks for a single-trip distance. Count the number of data points for segments A and B respectively, and determine whether the number of lost data points exceeds 10%. If it does, the flight must be restarted. If it meets the requirement, select the single-trip data segment with better data quality for analysis.

[0037] The azimuth obtained by radar detection is θ Rn The elevation angle is β Rn The azimuth obtained by photoelectric detection is θ Pn The elevation angle is β Pn The azimuth difference is denoted as △θ n The difference in elevation angle is denoted as △β. n , n∈[1,N];

[0038] △θ n =θ Rn -θ Pn

[0039] △β n =β Rn -β Pn

[0040] R∈[ts,ts*N]

[0041] Plot the distance-azimuth difference (R, Δθ) respectively. n Curve and distance-elevation difference (R, Δβ) n After handling singularities, the distance-azimuth difference curve is assumed to be as shown in the attached figure. Figure 4 As shown in the attached figure, the distance-elevation angle difference curve is as follows. Figure 5 As shown.

[0042] Based on the fluctuation trend, the distance-azimuth difference (R, Δθ) can be calculated. n The curve is divided into 3 segments, and 3 corresponding azimuth difference datasets are obtained. The number of data points in each segment is denoted as N. k The data sequence number within a segment is denoted as nk, and the difference in data between each segment is denoted as Δθ. nk Based on three azimuth difference datasets, according to The formula is used to obtain the mean azimuth difference △θ1, △θ2, and △θ3 corresponding to the three datasets, and the photoelectric azimuth is corrected for each distance segment by adding △θ1, △θ2, and △θ3 to the photoelectric azimuth data of the corresponding distance segment; the corrected photoelectric data is the actual detection data.

[0043] The distance-elevation angle difference (R, Δβ) n The curve is divided into four segments, and four corresponding elevation angle difference datasets are obtained. The number of data points in each segment is denoted as N. k The data sequence number within a segment is denoted as nk, and the difference in data between each segment is denoted as Δβ. nk Based on four elevation angle difference datasets, according to The formula is used to obtain the mean azimuth difference values ​​△β1, △β2, △β3, and △β4 corresponding to the four datasets, and the photoelectric elevation angle is corrected for each distance segment by adding △β1, △β2, △β3, and △β4 to the photoelectric elevation angle data of the corresponding distance segment; the corrected photoelectric data is the actual detection data.

[0044] Then, relaunch the UAV at an azimuth of 30° and record the radar and electro-optical detection data using the aforementioned method, and calculate the azimuth and elevation differences between the radar and electro-optical systems. Assuming the minimum field of view of the electro-optical system is 0.6°, the calibration for the current quadrant is complete when the azimuth error is less than 0.2° and the elevation error is less than 0.2°. Otherwise, the calibration needs to be repeated following the aforementioned steps.

[0045] Next, the calibration of quadrants (2), (3), (4), (5), and (6) will be carried out one by one according to the aforementioned method. By following this step, the calibration of the entire space will be completed, and the azimuth and elevation angle data of radar and electro-optical systems will be calibrated.

[0046] This invention focuses on a UAV detection and countermeasure system, specifically addressing the challenge of calibrating radar and optoelectronic devices and guiding them smoothly into the field of view within a 360-degree omnidirectional range when the mounting bases for radar and optoelectronic equipment cannot be perfectly leveled due to objective limitations. This method has been successfully implemented in engineering applications through testing.

Claims

1. A method for joint calibration of azimuth and elevation errors in a distributed radar optoelectronic system, characterized in that: Step 1: Divide the azimuth space where the radar and optoelectronic equipment are located into 6 quadrants; Step 2: Calibrate the data quadrant by quadrant. First, select the first quadrant for calibration, including: 1) Launch the UAV at the 30° azimuth of the middle position of the azimuth interval. The radar and electro-optical systems will respectively detect and obtain the target's azimuth and elevation angle values. The target's flight distance should be from 100 meters beyond the radar blind zone to 80% of the maximum detection power of the electro-optical system. Record two sets of data: one for the target far from the calibration equipment and one for the target near the calibration equipment. 2) Select one set of data from the two sets to statistically analyze the azimuth difference and elevation difference. The statistical method is as follows: record the radar track and electro-optical track data according to the radar rotation speed. Assume the radar rotation speed is ts, the number of data points recorded is N, and the azimuth acquired by the radar detection is θ. Rn The elevation angle is β Rn The azimuth obtained by photoelectric detection is θ Pn The elevation angle is β Pn The azimuth difference is denoted as △θ n The difference in elevation angle is denoted as △β. n , n∈[1,N]; △θ n =θ Rn -θ Pn △β n ＝β Rn -β Pn R∈[ts,ts*N] 3) Plot the distance-azimuth difference (R, Δθ) respectively. n Curve and distance-elevation difference (R, Δβ) n )curve; 4) Mark data points that differ from surrounding data by more than 0.5 as outliers, change the outlier values ​​to be the same as the previous data point, and redraw the distance-azimuth difference (R, Δθ). n Curve and distance-elevation difference (R, Δβ) n )curve; 5) Segment, statistically analyze, and correct the orientation data, including: observation distance - azimuth difference (R, Δθ) n The curve is used to segment the data according to its fluctuation trend. A total of N data points are divided into K segments, denoted as k, where k ∈ [1, K]. The number of data points in each segment is denoted as N. k The data sequence number within a segment is denoted as nk, and the azimuth difference data for each segment is denoted as Δθ. nk ; Calculate the mean azimuth difference of radar electro-optical signal within each segment. Assuming the normal azimuth detection data of photoelectric sensors at each distance segment within this quadrant are: The photoelectric orientation is corrected for each distance segment, and the corrected data is recorded as follows: 6) Segment, statistically analyze, and correct the elevation angle difference data, including: in the observation distance - elevation angle difference (R, Δβ) n The curve is used to segment the data according to its fluctuation trend. A total of N data points are divided into K segments, denoted as k, where k∈[1,K]. The number of data points in each segment is denoted as N. k The data sequence number within a segment is denoted as nk, and the difference in elevation angle for each segment is denoted as Δβ. nk The average elevation angle difference of the radar electro-optical system within each segment is calculated as follows: Assuming the normal elevation angle detection data of photoelectric sensors for each distance segment within this quadrant are: The elevation angle of the photoelectric sensor is corrected for each distance segment, and the corrected data is recorded as follows: Step 3: Fly another flight, and follow steps 1) to 3) in Step 2 to calculate the azimuth and elevation errors between the radar and electro-optical systems, and calculate the mean azimuth and elevation errors. If the mean azimuth error is less than one-third of the electro-optical azimuth field of view, and the mean elevation error is less than one-third of the electro-optical azimuth field of view, the calibration of the current quadrant ends and proceeds to Step 4; otherwise, proceed to Step 2. Step 4: Launch the drone at the midpoint of each quadrant from 2 to 6, and repeat steps 2 and 3 to complete the data measurement, calculation and correction for the current quadrant.

2. The method for joint calibration of azimuth and elevation errors in a distributed radar optoelectronic system according to claim 1, characterized in that: The spatial range is divided into six quadrants in azimuth, and each quadrant is corrected according to the same set of correction values. Taking the center of the line connecting the photoelectric and radar as the origin, and the end where the photoelectric and radar rays are located as the azimuth 0 position, the entire azimuth plane is divided into six quadrants in a counterclockwise direction with 60° as one quadrant, forming six quadrants in total: Quadrant 1: 0-60°, Quadrant 2: 60°-120°, Quadrant 3: 120°-180°, Quadrant 4: 180°-240°, Quadrant 5: 240°-300°, and Quadrant 6: 300°-360°.

3. The method for joint calibration of azimuth and elevation errors in a distributed radar optoelectronic system according to claim 1, characterized in that: In the second step, the methods for selecting one set of data from the two sets include: Select data with a radar track loss rate of less than 10% and an electro-optical track loss rate of less than 10% for use. If neither can be achieved, the UAV needs to be re-launched to acquire data. The track loss rate is calculated as follows: the number of data points that should be in the flight distance segment is denoted as N_all, the number of track data points actually recorded is N_Track, and the track loss rate is Track-Loss-Rate: Track-Loss-Rate = (N_all - N_Track) / N_all.

4. The method for joint calibration of azimuth and elevation errors in a distributed radar optoelectronic system according to claim 1, characterized in that: In the second step, the distance is divided into segments according to the distribution of the distance-azimuth difference curve and the distance-elevation difference. The data with a azimuth difference or elevation difference fluctuation exceeding 0.3 degrees is divided into segments.

Images