A multi-track fusion method

Abstract

The application discloses a multi-track fusion method, which comprises the following steps: a radar A detects a known target to obtain a first track composed of a plurality of first point tracks; a radar B detects the same known target to obtain a second track composed of a plurality of second point tracks; according to the first track and the second track, a first error correction value for correcting the first track of the radar A and a second error correction value for correcting the second track of the radar B are determined under the constraint condition that the first track and the second track are coincident; and the first track and the second track are respectively corrected by using the first error correction value and the second error correction value, so that a track is obtained by fusing the first track and the second track. The method can realize track fusion of the same air target, so that whether the multi-track is the track of the same target or the track of different targets can be effectively distinguished.

Description

Technical Field

[0001] This invention relates to information fusion technology, and in particular to a method for multi-track fusion. Background Technology

[0002] When an aerial target is detected by more than one radar, each radar will provide the target's spatial location. Due to inconsistencies in detection time and errors in detection range and azimuth, each radar provides a different spatial location, making it difficult to distinguish between the same and different targets. In track fusion, this significant error prevents the application from achieving its intended purpose. Summary of the Invention

[0003] This invention provides a method for multi-track fusion, which solves the problem that existing technologies cannot fuse multiple tracks of the same aerial target.

[0004] This invention provides a method for multi-track fusion, comprising the following steps: Radar A detects a known target and obtains a first track composed of multiple first points, each first point including a point position and a point time; Radar B detects the same known target and obtains a second track composed of multiple second points, each second point including a point position and a point time; based on the point positions and times of the first track and the second track, a first error correction value for Radar A to correct the first track and a second error correction value for Radar B to correct the second track are determined, with the overlap of the first track and the second track as a constraint; the first error correction value and the second error correction value are used to correct the first track and the second track respectively, thereby obtaining a track fused from the first track and the second track.

[0005] Preferably, the step of determining the first error correction value for radar A to correct the first track and the second error correction value for radar B to correct the second track based on the position and time of the first track and the position and time of the second track, with the overlap of the first track and the second track as a constraint, includes: when the position and time of the first track are not synchronized with the position and time of the second track, obtaining the point PB1′(T1′, Rb1′, Ξb1′) of each target point time T1′ of the second track that is the same as the target point time T2 of the first track; and based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Y) of the respective base points of radar A and radar B, respectively. b, Zb), for each point PA1(T2, Ra1, Ξa1) and each point PB1′(T1′, Rb1′, Ξb1′) of the first track, a mathematical model for error correction is determined; based on the mathematical model for error correction, a first error value for each target point PA1(T2, Ra1, Ξa1) of radar A and a second error value for each point PB1′(T1′, Rb1′, Ξb1′) of radar B are obtained; the convergence value of the first error values ​​of the multiple target points of radar A is used as the first error correction value for radar A to correct the first track; the convergence value of the second error values ​​of the multiple target points of radar B is used as the second error correction value for radar B to correct the second track.

[0006] Preferably, the step of determining the first error correction value for radar A to correct the first track and the second error correction value for radar B to correct the second track based on the position and time of the first track and the position and time of the second track, with the overlap of the first track and the second track as a constraint, further includes: when the position and time of the first track are synchronized with the time of the second track, setting the point PB1′(T2, Rb1′, Ξb1′) of each target point time T2 of the second track that is the same as the time of each target point time T2 of the first track.

[0007] Preferably, the step of obtaining the trace PB1′(T1′, Rb1′, Ξb1′) of each target trace time T1′ of the second track that is the same as each target trace time T2 of the first track includes: based on each target trace time T2 of the first track, determining the trace times T1 and T3 of the second track that are adjacent to each target trace time T2 of the first track; obtaining the trace PB1(T1, Rb1, Ξb1) and trace PB2(T3, Rb2, Ξb2) of the second track according to the trace times T1 and T3; and calculating the trace PB1′(T1′, Rb1′, Ξb1′) of the second track according to the trace PB1(T1, Rb1, Ξb1) and trace PB2(T3, Rb2, Ξb2).

[0008] Preferably, the step of determining the mathematical model for error correction based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of radar A and radar B, and the traces PA1 (T2, Ra1, Ξa1) and PB1′ (T1′, Rb1′, Ξb1′) of the first track includes: obtaining the vertical projection points C and O of radar A and the trace PA1 on the horizontal plane where radar B is located, respectively, so that the vertical projection points C, O, and radar B form a triangle ΔCBO; based on the three-dimensional coordinates (Xa, Ya, Za) of the base points of radar A and radar B, the mathematical model for error correction includes: obtaining the vertical projection points C and O of radar A and radar B on the horizontal plane where radar B is located, respectively, so that the vertical projection points C, O, and radar B form a triangle ΔCBO; and determining the mathematical model for error correction based on the three-dimensional coordinates (Xa, Ya, Za) of the base points of radar A and radar B. Based on the azimuth angles Ξa1 and Ξb1′ of radar A and radar B relative to the known target at target trace time T2, the distance D of CB in ΔCBO, the angle α of ∠OCB, and the angle β of ∠OBC are determined; according to D, α, β, the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of radar A and radar B, the distances Ra1 and Rb1′ between radar A and radar B and the known target at target trace time T2, a mathematical model for error correction is determined; the mathematical model for error correction is: X=[D 2 +Ra1 2 -(H-Za+Zb) 2 -Rb1′ 2 +H 2 ] / 2D;Ra1 2 +H 2 +D 2 (tgα-tgβ) / (tgα+tgβ)-(H-Za+Zb) 2 -Rb1′ 2=0; where H is the preset height of the known target relative to the horizontal plane where ΔCBO is located; X is the straight-line distance between point C in ΔCBO and the perpendicular foot point E, and E is the perpendicular foot of point O in ΔCBO at CB.

[0009] Preferably, determining the distance D of CB, the angle α of ∠OCB, and the angle β of ∠OBC in ΔCBO based on the base point three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of radar A and radar B, and the azimuth angles Ξa1 and Ξb1′ of radar A and radar B relative to the known target at target trace time T2 includes: determining the distance D of CB in ΔCBO based on the base point three-dimensional coordinates (Xa, Ya, Za) of radar A and the base point three-dimensional coordinates (Xb, Yb, Zb) of radar B; determining the angle α of ∠OCB in ΔCBO based on Ξa1 and the base point three-dimensional coordinates (Xb, Yb, Zb) of radar B; and determining the angle β of ∠OBC in ΔCBO based on Ξb1′ and the base point three-dimensional coordinates (Xb, Yb, Zb) of radar B.

[0010] Preferably, the first error value of each target point PA1(T2, Ra1, Ξa1) of radar A includes a relative angle correction Δα0, and the second error value of each point PB1′(T1′, Rb1′, Ξb1′) of radar B includes a relative angle correction Δβ0; obtaining the first error value of each target point PA1(T2, Ra1, Ξa1) of radar A and the second error value of each point PB1′(T1′, Rb1′, Ξb1′) of radar B according to the mathematical model for error correction includes: according to the mathematical model for error correction, the error value is obtained from the vertical projection point. C. Using the cosine trigonometric function relationship in the right triangle ΔCEO formed by the vertical projection point O and the perpendicular foot point E, the theoretical angle α0 of ∠OCE of the known target at the preset height is obtained; based on the mathematical model used for error correction and the cosine trigonometric function relationship in the right triangle ΔBEO formed by the radar B, the vertical projection point O, and the perpendicular foot point E, the theoretical angle β0 of ∠OBE of the known target at the preset height is obtained; based on α and α0, the relative angle correction amount Δα0 of the radar A is determined; based on β and β0, the relative angle correction amount Δβ0 of the radar B is determined.

[0011] Preferably, the first error correction value for radar A to correct the first track includes an angle correction amount Δa, and the second error correction value for radar B to correct the second track includes an angle correction amount Δβ. The step of using the first error correction value and the second error correction value to correct the first track and the second track respectively, thereby obtaining a track formed by merging the first track and the second track, includes: using Δa to correct multiple first points constituting the first track to obtain a corrected first track, and using Δβ to correct multiple second points of the second track to obtain a corrected second track, thereby obtaining a track formed by merging the first track and the second track.

[0012] This invention uses the error correction value of each radar to correct the track, and performs position correction on the track (or the point trace that forms the track) detected by each radar, so as to realize the fusion of multiple tracks for the same air target and effectively distinguish whether the multiple tracks are tracks of the same target or tracks of different targets. Attached Figure Description

[0013] Figure 1 This is a flowchart of the multi-track fusion method provided by the present invention;

[0014] Figure 2 This is a schematic diagram of two radars detecting targets provided by the present invention;

[0015] Figure 3 This is a flowchart for determining the first error correction value and the second error correction value. Detailed Implementation

[0016] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In the following description, suffixes such as "module," "part," or "unit" used to denote elements are used only for the purpose of illustrative purposes and have no inherent meaning. Therefore, "module," "part," or "unit" may be used interchangeably.

[0017] For a single target in the air, there is only one flight path. However, for different radars that detect this target, due to the inconsistent detection time and errors in detection range and azimuth, each radar gives a different spatial position, resulting in different flight paths. Under a unified coordinate system, this invention determines the error of each radar based on the uniqueness of the airborne target. Specifically, assuming there is one target in the air (in actual use, this can be manually set according to conditions, such as secondary response confirmation for civil aviation flights), if there are multiple detection radars, ten batches of message information (ten flight paths) will be generated. Whether these ten batches of message information represent one target or ten targets is unclear to the information processing system. This invention, based on the known case of a single target, determines under what conditions the ten batches of message information can be identified as belonging to the same target; this condition is the radar error to be found.

[0018] See Figure 1 The multi-track fusion method of the present invention may include the following steps:

[0019] Step S101: Radar A detects a known target and obtains a first track consisting of multiple first points, each first point including the point position and point time; Radar B detects the same known target and obtains a second track consisting of multiple second points, each second point including the point position and point time.

[0020] The known target is an aerial target, such as a flying vehicle or aircraft.

[0021] The location of the dot includes the distance between the radar and the target and the azimuth angle of the target detected by the radar.

[0022] Step S102: Based on the position and time of the first track and the position and time of the second track, determine the first error correction value for radar A to correct the first track and the second error correction value for radar B to correct the second track, with the overlap of the first track and the second track as a constraint.

[0023] Step S103: Correct the first track and the second track using the first error correction value and the second error correction value respectively, thereby obtaining a track formed by merging the first track and the second track.

[0024] The method of this invention uses radar to achieve track fusion. For example, radar B sends a detected second track to radar A. Radar A then executes steps S102 and S103 based on its own detected first track and the received second track from radar B to achieve track fusion for the same target (e.g., the same aircraft). Similarly, radar A can also send its detected first track to radar B. In this way, radar B can execute steps S102 and S103 based on its own detected second track and the received first track from radar A to achieve track fusion for the same target (e.g., the same aircraft).

[0025] The apparatus of the present invention will be described in detail below through two embodiments.

[0026] Example 1

[0027] Assuming that both radar A and radar B detect target P, due to errors in detection range and azimuth between radar A and radar B, two different tracks appear: the first track of radar A detecting target P and the second track of radar B detecting target P.

[0028] A first error correction value (ΔRa, Δα) is set for the trajectory correction of radar A, and a second error correction value (ΔRb, Δβ) is set for the trajectory correction of radar B.

[0029] The first error correction value (ΔRa, Δα) and the second error correction value (ΔRb, Δβ) are continuously adjusted until it can be determined that the two tracks after correction based on the first error correction value (ΔRa, Δα) and the second error correction value (ΔRb, Δβ) are merged into one track of the target P. At this point, the first error correction value (ΔRa, Δα) and the second error correction value (ΔRb, Δβ) are respectively used as error correction values ​​that can be applied in practice.

[0030] After using the practically applicable error correction values ​​of radar A and radar B to correct multiple points on the first track of target P detected by radar A and multiple points on the second track of target P detected by radar B, the two tracks can be merged into one track.

[0031] When multiple tracks (e.g., 10) of the same or different targets are detected, the tracks of the targets detected by each radar can be corrected by using the error correction values ​​available for practical application of each radar. Track fusion of the same targets can be achieved, thereby distinguishing between the same targets and different targets.

[0032] Example 2

[0033] In one implementation, see Figure 2Taking two radars, A and B, as an example, under a unified coordinate system, the three-dimensional coordinates of the base points of radar station A and radar station B are respectively: A(Xa, Ya, Za) and B(Xb, Yb, Zb).

[0034] Suppose radar A detects multiple points on a known target P, forming a track, denoted as the first track. See [link to relevant documentation]. Figure 2 The at least two target points of the first track are PA1(T2, Ra1, Ξa1) and PA2(T4, Ra2, Ξa2).

[0035] Assume radar B detects multiple points of the same target P, forming a track, denoted as the second track. See [link / reference]. Figure 2 The at least two target points of the second track are PB1(T1, Rb1, Ξb1) and PB2(T3, Rb2, Ξb2).

[0036] Where T(T1, T2, T3, T4) is the time when the spot was generated (referred to as the spot time), R(Ra1, Ra2, Rb1, Rb2) is the distance between the radar and the target, and Ξ(Ξa1, Ξa2, Ξb1, Ξb2) is the azimuth angle of the target.

[0037] See Figure 3 The steps of determining a first error correction value for radar A to correct a first track and a second error correction value for radar B to correct a second track include:

[0038] Step S301: When the time of the first track is not synchronized with the time of the second track, calculate the point PB1′(T1′, Rb1′, Ξb1′) of each target point time T1′ of the second track that is the same as the time of each target point T2 of the first track.

[0039] See Figure 2Based on the target point time T2 of the first track, determine the point times T1 and T3 of the second track that are adjacent to the target point time T2 of the first track. Then, according to the point times T1 and T3, obtain the point PB1(T1, Rb1, Ξb1) and point PB2(T3, Rb2, Ξb2) of the second track. And according to the point PB1(T1, Rb1, Ξb1) and point PB2(T3, Rb2, Ξb2) of the second track, calculate the point PB1′(T1′, Rb1′, Ξb1′) of the second track. For example, the following steps can be used to calculate: the distance change after the target moves is ΔR=[(Rb2-Rb1) / (T3-T1)](T2-T1); the angle change after the target moves is ΔΞ=[(Ξb2-Ξb1) / (T3-T1)](T2-T1). Thus, Rb1′=Rb1+ΔR; Ξb1′=Ξb1+ΔΞ; T1′=T2. From this, we obtain the extrapolated point (or trace) PB1′(T1′,Rb1′,Ξb1′) at time T1′.

[0040] It should be noted that Rb1′ is the extrapolation point of radar B on PB1, and due to errors, it is impossible for it to coincide with PA1 in reality.

[0041] In another embodiment, when the time of the first track is synchronized with the time of the second track, for ease of explanation, the track PB1′(T2, Rb1′, Ξb1′) of each target track time T2 of the second track is set to be the same as that of each target track time T2 of the first track.

[0042] Step S302: Determine the mathematical model for error correction based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of radar A and radar B, and each point PA1 (T2, Ra1, Ξa1) and each point PB1′ (T1′, Rb1′, Ξb1′) of the first track.

[0043] See Figure 2 Plane CBO is parallel to the horizontal plane. Point C is the vertical projection of radar station A onto the horizontal plane. Point O is the projection of the aerial target onto the horizontal plane at time T2. Point E is the foot of the perpendicular from point O to CB in ΔCBO. AO Let L be the projection of Ra1 onto the CBO plane; BOLet Rb1′ be the projection of the target onto the CBO plane, H be the height of the target relative to the CBO plane, N be the north direction, X be the length of line CE, D be the length of line CB (which is actually equal to the straight-line distance between radar stations A and B), α be the angle of ∠OCB, β be the angle of ∠OBC, HOE be the height of ΔCOB relative to point O, point A be the location of radar station A, and point B be the location of radar station B.

[0044] First, obtain the vertical projection point C of radar A on the horizontal plane where radar B is located, and the vertical projection point O of the aerial target's trace PA1 on the horizontal plane where radar B is located. Thus, the vertical projection point C, the vertical projection point O, and radar B form a triangle ΔCBO.

[0045] Then, based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of radar A and radar B, and the azimuth angles Ξa1 and Ξb1′ of radar A and radar B relative to the known target at target trace time T2, the distance D of CB in ΔCBO, the angle α of ∠OCB, and the angle β of ∠OBC are determined.

[0046] Specifically, the distance D of CB in ΔCBO is determined based on the three-dimensional coordinates (Xa, Ya, Za) of the base point of radar A and the three-dimensional coordinates (Xb, Yb, Zb) of the base point of radar B; the angle α of ∠OCB in ΔCBO is determined based on Ξa1 and the three-dimensional coordinates (Xb, Yb, Zb) of the base point of radar B; and the angle β of ∠OBC in ΔCBO is determined based on Ξb1′ and the three-dimensional coordinates (Xb, Yb, Zb) of the base point of radar B.

[0047] Assuming radar A is the origin of the coordinate system, i.e., Xa = 0, Ya = 0, Za = 0, then we can obtain:

[0048] D=(Xb 2 +Yb 2 ) 1 / 2 ;

[0049] α=180°-Ξa1-arctg|Xb / Yb|;

[0050] β=Ξb1′+arctg|Xb / Yb|;

[0051] Finally, based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of D, α, β, radar A and radar B respectively, the distances Ra1 and Rb1′ between radar A and radar B and the known target at target trace time T2, a mathematical model for error correction is determined.

[0052] Depend on Figure 2 From ΔEOC and ΔEOB, we know that Xtgα=(DX)tgβ, and X can be solved as shown in formula (1).

[0053] X=Dtgβ / (tgα+tgβ) (1)

[0054] Depend on Figure 2 From ΔEOB, we know that:

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

[0061] Therefore, the following relationship (2) between H and Ra1, α, and Rb1′β can be solved:

[0062] X = [D] 2 +Ra1 2 -(H-Za+Zb) 2 -Rb1′ 2 +H 2 ) / 2D (2)

[0063] From equations (1) and (2), we can obtain:

[0064] Dtgβ / (tgα+tgβ)=[D 2 +Ra1 2 -(H-Za+Zb) 2 -Rb1′ 2 +H 2 ) / 2D;

[0065] D 2 +Ra1 2 -(H-Za+Zb) 2 -Rb1′ 2 +H 2 -2D 2 tgβ / (tgα+tgβ)=0;

[0066] Ra1 2 -(H-Za+Zb) 2 -Rb1′ 2 +H 2 +D 2[1-2tgβ / (tgα+tgβ)]=0;

[0067] Ra1 2 -(H-Za+Zb) 2 -Rb1′ 2 +H 2 +D 2 (tgα-tgβ) / (tgα+tgβ)=0;

[0068] Ra1 2 +H 2 +D 2 (tgα-tgβ) / (tgα+tgβ)-(H-Za+Zb) 2 -Rb1′ 2 =0 (3)

[0069] Equation (3) finally establishes the equations for H and the known quantities Ra1, α, and Rb1′β.

[0070] When two planar radars can detect a group of targets simultaneously, the altitude of the aerial targets can be calculated from the data obtained by the two radars.

[0071] It should be noted that the mathematical model for error correction in this invention is based on the following conditions: (1) Radar station A and radar station B must provide relatively accurate three-dimensional coordinates of the base point; (2) When processing the data of the two radar stations, it must be clearly known that the data being processed is the same batch of targets; (3) The recording should have a unified time calibration, and the error after calibration should be less than 1 second. Time error will affect the credibility of the radar measured data and reduce the credibility of the calculated results; (4) It is best to use automatic recording to avoid random human error.

[0072] Step S303: Based on the mathematical model used for error correction, obtain the first error value of each target point PA1(T2, Ra1, Ξa1) of radar A and the second error value of each point PB1′(T1′, Rb1′, Ξb1′) of radar B.

[0073] The theoretical basis for the validity of the mathematical model for error correction in this invention includes:

[0074] (1) The uniqueness of space targets.

[0075] Theoretically, if two radars measure target data simultaneously without error, then there exists a unique point in space. The main reason for point separation is that the distance and azimuth measured by the radars have different magnitudes of error, and the difference between the actual point position and the measured point position is the radar error. When establishing the mathematical model, we first assume that the radar has no error, then there exists a unique point in space, and this point is the true position of the target, so that all data can satisfy equation (3). Conversely, under the premise of no measurement error, equation (3) also restricts the uniqueness of the target's spatial point.

[0076] (2) The special characteristics of radar error

[0077] The range error of a two-coordinate radar is caused by the accuracy of echo signal processing, exhibiting relative stability, and the target position shift caused by the range error is negligible. The azimuth error of a two-coordinate radar is caused by true north error and the mechanical rotation of the radar antenna, with true north error being the most significant. Whether azimuth or range, the error always consists of two parts: a fixed error and a random error. This can be expressed as: ΔR = r + ε r And ΔΞ=θ+ε θ Where ΔR is the total distance error; r is the fixed distance error; ε r ε is the random distance error; ΔΞ is the total azimuth error; θ is the fixed azimuth error; ε θ The azimuth random error is ΔR. ΔR has little impact on the true position of the target. Furthermore, the error can be transferred using equation (3) (i.e., ΔR error is ignored, only the error of ΔΞ is considered, so ΔΞ is increased by the increment caused by the ΔR error on top of the original error). Considering the azimuth uniformly, ΔR can be ignored. The error that has the greatest impact on the true position of the target comes from ΔΞ, and the main component of ΔΞ is θ. The characteristics of θ are relatively fixed. As long as θ can be corrected, and the transfer of the ΔR error is considered when correcting θ, according to ε... r and ε θ The change in can be calculated using equation (3), which can give the approximate range of the error of the solution value under given conditions.

[0078] Based on the above, the first error value of each target point PA1(T2, Ra1, Ξa1) of radar A may include a relative angle correction Δα0, and the second error value of each target point PB1′(T1′, Rb1′, Ξb1′) of radar B may include a relative angle correction Δβ0. Specifically, according to the mathematical model used for error correction and the cosine trigonometric function relationship in the right triangle ΔCEO formed by the vertical projection point C, the vertical projection point O, and the perpendicular foot point E, the theoretical angle α0 of ∠OCE of the known target at the theoretical height H0 is obtained; according to the mathematical model used for error correction and the cosine trigonometric function relationship in the right triangle ΔBEO formed by radar B, the vertical projection point O, and the perpendicular foot point E, the theoretical angle β0 of ∠OBE of the known target at the theoretical height H0 is obtained; the relative angle correction Δα0 of radar A is determined according to α and α0; and the relative angle correction Δβ0 of radar B is determined according to β and β0.

[0079] See Figure 2 Ra1, α, Rb1′, β, Za, Zb, and D are known quantities that can be calculated.

[0080] Set an initial value for H0, for example, H0 = 6 km, and calculate α0 and β0, where α0 is the theoretical angle of the target relative to radar A when the altitude is equal to H0, and β0 is the theoretical angle of the target relative to radar B when the altitude is equal to H0. The value of H will not have a significant impact on error correction.

[0081] according to From equations (2) and (3), we can obtain the following equation:

[0082]

[0083] Therefore, at height H0, we can obtain

[0084]

[0085] according to From equations (2) and (3), we can obtain the following equation:

[0086]

[0087] Therefore, at height H0, we can obtain

[0088]

[0089] In this way, the relative angle corrections Δα0 and Δβ0 can be obtained.

[0090] Δα0=α0-α (4)

[0091] Δβ0=β0-β (5)

[0092] By correcting the tracks (or azimuth angles in the measured track data) of radar A and radar B according to Δα0 and Δβ0 respectively, the data tracks of the two radars can be made to overlap.

[0093] Step S304: The convergence value of the first error value of the multiple target point traces of radar A is used as the first error correction value for radar A to correct the first trajectory, and the convergence value of the second error value of the multiple target point traces of radar B is used as the second error correction value for radar B to correct the second trajectory.

[0094] To improve correction accuracy, multiple first error values ​​are obtained for multiple target point tracks of radar A. Then, a convergence value for the multiple first error values ​​of the multiple target point tracks of radar A is determined, and this convergence value is used as the first error correction value Δα for radar A to correct its first trajectory. Similarly, multiple second error values ​​are obtained for multiple target point tracks of radar B. Then, a convergence value for the multiple second error values ​​of the multiple target point tracks of radar B is determined, and this convergence value is used as the second error correction value Δβ for radar B to correct its second trajectory.

[0095] It should be noted that, similar to the north-pointing calibration test, an accurate H value can be obtained through the test flight for error correction. The initial H value is also set considering the target distance, aiming to be as close to the actual situation as possible, thus achieving higher correction accuracy.

[0096] After obtaining the first error correction value Δα for radar A to correct the trajectory and the second error correction value Δβ for radar B to correct the trajectory, the multiple first points constituting the first trajectory can be corrected using Δα to obtain the corrected first trajectory, and the multiple second points of the second trajectory can be corrected using Δβ to obtain the corrected second trajectory, thereby obtaining a trajectory formed by merging the first trajectory and the second trajectory.

[0097] Using the above method, when multiple tracks (e.g., 10) of the same or different targets appear, the track of each radar is corrected by its own error correction value, thereby achieving track fusion of the same target and distinguishing between the same target and different targets.

[0098] This invention achieves multi-track fusion for the same aerial target by performing position correction on each radar-detected track (or the point trace forming the track), effectively distinguishing whether the multiple tracks are for the same target or for different targets.

[0099] The preferred embodiments of the present invention have been described above with reference to the accompanying drawings, but this does not limit the scope of the invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and spirit of the present invention should be within the scope of the present invention.

Claims

1. A method for multi-track fusion, characterized in that, The method includes the following steps: Radar A detects a known target and obtains a first track consisting of multiple first points, each first point including the point position and point time; Radar B detects the same known target and obtains a second track consisting of multiple second points, each of which includes the position and time of the point; Based on the position and time of the first track and the position and time of the second track, and with the overlap of the first and second tracks as a constraint, a first error correction value for radar A to correct the first track and a second error correction value for radar B to correct the second track are determined, including: When the time of the first track is not synchronized with the time of the second track, calculate the track PB1′ (T1′,Rb1′,Ξb1′) of each target track time T1′ of the second track that is the same as the target track time T2 of the first track. Based on the three-dimensional coordinates (Xa,Ya,Za) and (Xb,Yb,Zb) of the base points of radar A and radar B, and each point PA1 (T2,Ra1,Ξa1) and each point PB1′ (T1′,Rb1′,Ξb1′) of the first track, a mathematical model for error correction is determined. Based on the mathematical model used for error correction, the first error value of each target point PA1 (T2, Ra1, Ξa1) of radar A and the second error value of each point PB1′ (T1′, Rb1′, Ξb1′) of radar B are obtained. The convergence value of the first error values ​​of the multiple target point tracks of radar A is used as the first error correction value for radar A to correct the first track; The convergence value of the second error values ​​of the multiple target point tracks of radar B is used as the second error correction value for radar B to correct the second trajectory; The first track and the second track are corrected using the first error correction value and the second error correction value respectively, thereby obtaining a track formed by merging the first track and the second track.

2. The method according to claim 1, characterized in that, The step of determining the first error correction value for radar A to correct the first track and the second error correction value for radar B to correct the second track, based on the position and time of the first track and the position and time of the second track, with the overlap of the first track and the second track as a constraint, further includes: When the time of the first track is synchronized with the time of the second track, the track PB1′ (T2,Rb1′,Ξb1′) of each target track time T2 of the second track is set to be the same as that of each target track time T2 of the first track.

3. The method according to claim 1, characterized in that, The step of obtaining the trace PB1′ (T1′, Rb1′, Ξb1′) of the second track, which has the same target trace time T2 as the first track, for each target trace time T1′, includes: Based on the target point time T2 of the first track, determine the point times T1 and T3 in the second track that are adjacent to the target point time T2 of the first track; Based on the time points T1 and T3, obtain the points PB1 (T1, Rb1, Ξb1) and PB2 (T3, Rb2, Ξb2) of the second track. Based on the points PB1(T1,Rb1,Ξb1) and PB2(T3,Rb2,Ξb2) of the second track, the point PB1′(T1′,Rb1′,Ξb1′) of the second track is calculated.

4. The method according to any one of claims 2-3, characterized in that, The step of determining the mathematical model for error correction based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of radar A and radar B, and the points PA1 (T2, Ra1, Ξa1) and PB1′ (T1′, Rb1′, Ξb1′) of the first track includes: The vertical projection points C and O of radar A and the point PA1 on the horizontal plane where radar B is located are obtained respectively, so that the vertical projection point C, the vertical projection point O and radar B form a triangle ΔCBO. Based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of radar A and radar B, and the azimuth angles Ξa1 and Ξb1′ of radar A and radar B relative to the known target at target trace time T2, determine the distance D of CB in ΔCBO, the angle α of ∠OCB, and the angle β of ∠OBC. Based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of D, α, β, radar A, and radar B, respectively, and the distances Ra1 and Rb1′ between radar A and radar B and the known target at target trace time T2, a mathematical model for error correction is determined; the mathematical model for error correction is: ； ； Where H is the preset height of the known target relative to the horizontal plane where ΔCBO is located; X is the straight-line distance between point C in ΔCBO and the perpendicular foot point E, and E is the perpendicular foot of point O in ΔCBO at CB.

5. The method according to claim 4, characterized in that, The step of determining the distance D of CB in ΔCBO, the angle α of ∠OCB, and the angle β of ∠OBC based on the three-dimensional coordinates (Xa, Ya, Za) and (Xb, Yb, Zb) of the base points of radar A and radar B, and the azimuth angles Ξa1 and Ξb1′ of radar A and radar B relative to the known target at target trace time T2 includes: Based on the three-dimensional coordinates (Xa, Ya, Za) of the base point of radar A and the three-dimensional coordinates (Xb, Yb, Zb) of the base point of radar B, determine the distance D of CB in ΔCBO; Based on the three-dimensional coordinates (Xb, Yb, Zb) of the base point of Ξa1 and the radar B, determine the angle of ∠OCB in ΔCBO. ; Based on the three-dimensional coordinates (Xb, Yb, Zb) of the base point of Ξb1′ and the radar B, determine the angle of ∠OBC in ΔCBO. .

6. The method according to claim 5, characterized in that, The first error value for each target point PA1 (T2, Ra1, Ξa1) of radar A includes a relative angle correction. The second error value for each point PB1′ (T1′, Rb1′, Ξb1′) of radar B includes a relative angle correction. The step of obtaining the first error value of each target point PA1(T2,Ra1,Ξa1) of radar A and the second error value of each point PB1′(T1′,Rb1′,Ξb1′) of radar B according to the mathematical model used for error correction includes: Based on the mathematical model used for error correction and the cosine trigonometric function relationships in the right triangle ΔCEO formed by the vertical projection point C, the vertical projection point O, and the foot of the perpendicular E, the theoretical angle ∠OCE of the known target at the preset height is obtained. ; Based on the mathematical model used for error correction and the cosine trigonometric function relationships in the right triangle ΔBEO formed by the radar B, the vertical projection point O, and the perpendicular foot point E, the theoretical angle ∠OBE of the known target at the preset height is obtained. ; According to the above and stated Determine the relative angle correction amount of radar A. ; According to the above and stated Determine the relative angle correction amount of radar B. .

7. The method according to claim 6, characterized in that, The first error correction value for radar A to correct the first trajectory includes the angle correction amount. The second error value for radar B to correct the second trajectory includes an angle correction amount. The step of correcting the first track and the second track using the first error correction value and the second error correction value respectively, thereby obtaining a track formed by merging the first track and the second track, includes: Using the The multiple first points constituting the first track are corrected to obtain the corrected first track, and the corrected first track is obtained by utilizing the... The second track is modified by correcting multiple second points to obtain a modified second track, thus obtaining a track formed by merging the first track and the second track.

Images