JPS61205883A - Apparatus for tracking moving target - Google Patents

Abstract

PURPOSE:To enable tracking without delay even if a moving target takes acceleration motion, by increasing the weighing coefficient of a tracking filter by the difference between speed vectors calculated from two radars arranged so as to be spaced apart by a predetermined distance. CONSTITUTION:Doppler frequencies are respectively calculated on the basis of transmission frequency of a transmitter 7 and the receiving frequencies of receivers 8, 9 by Doppler frequency detectors 10, 11. Subsequently, a speed, wherein the speed vector of a moving target is projected to the straight lines connecting antennae 1, 2 and the target, is calculated by a beam axis speed detectors 15, 16 and the X-axis and Y-axis direction components of the speed of the target are calculated by a target speed vector calculator 17. A target position calculator 12 converts the data of the polar coordinates system from radar to the data of the orthogonal coordinates system. Next, the difference between the smoothed speed vector of a tracking filter 18 and the speed vector of the calculator 17 is calculated by a speed vector judge device 19 through the filter 18 and, when the difference is larger than a tolerant error, it is judged that the moving target takes accerelation motion, weighing coefficient is increased and the estimation error of a future position can be reduced.

Description

[Detailed description of the invention] [Industrial application field] This emission 814 is generated by the first radar and the second radar.

By evaluating the velocity vector calculated by observing the Doupler frequency of a moving target such as an aircraft and the velocity vector of the moving target obtained in the tracking filter, it is possible to efficiently track the accelerated movement of the moving target. This invention relates to a moving target tracking device.

[Conventional technology]

With the conventional device, the position kt of a moving target such as an aircraft observed by radar, the polar coordinate RD with the radar as the origin,
Convert from OP to the following formula il1, formula (2) (XD*YD) in the orthogonal coordinate system,
In addition to smoothing the position of the moving target using equation 1,
After smoothing the multivelocity using equations 5) and 6, we obtain (7
)formula.

The future position of the moving target is fully predicted using equation (8).

In parallel with this, as the weighting coefficients α and βin increase, they are repeatedly monotonically decreased.

XD (n) -RD (n) ・8in 0D(n)
(11YD(n) = Rp (
n) -cos Op (n) (
21Xs (n) = Xp (n) + α(XD (n
) −Xp (n) ) (31YB
(n)-Yp (n)+α(YD (n)-Yp (n
) ) +4+X(n)=X(n-1
)+' (Xp(n)−Xp(n))
! 51Y(n)-Y(n-1)+'(YD
(n)-Yp(n)) i61XP(n
-1-Lee-XB(n)+T-M(n)
f71Yp (n+Lee-YB(n)+T-Y
(n) (81εx(rl-+
-Li XD n+t -Xp(n+-Li)2
f91εy(n+su! YD n+1
−Yp(n+su)2 αυε(n+suIl!
Isr1+1+ty(n+1)2(1υjust 11~
In equation (8), RD and OD are the observed values of the position of the moving target by the radar in the polar coordinate system.

XD * YD is the value converted to RD, OD k orthogonal coordinate system.

XB*YB is the smoothed value of the target position, X, ÷ is the smoothed value of the velocity, xp, Yp are the predicted values of the target position, α is the weighting coefficient for position smoothing, and β is the weighting coefficient for velocity smoothing.

T represents the data sampling interval, and n represents the sampling time.

[Problem that the invention seeks to solve]

In this conventional device, smoothing of the target position and velocity is performed by linear calculation of the position, so α, β
If the target performs acceleration movement bt- after it has sufficiently decreased, the errors in the target's predicted position, smooth speed, and direction of movement will increase, and as a result, the error function of the υ equation will become large and the value of ε(n) will increase. There was a problem in that accelerated motion could only be detected after the size had increased.

The present invention has been made to overcome all of these problems, and aims to track a moving target without delay even when the moving target shifts from uniform linear motion to accelerated motion.

[Means for solving problems]

This invention provides means for calculating the entire velocity vector of a target at a sampling time when the target is observed, using a first radar and a second radar installed at a predetermined distance apart, and the Doupler frequency obtained by these two radars. , with the velocity vector above.

and means for calculating a vector difference between the smoothed velocity vector and the smoothed velocity vector obtained by linear calculation of the position.

[Effect]

In this invention, the acceleration motion of the y16 moving target is detected based on the difference between the two calculated velocity vectors, and the weighting coefficients α and β of the tracking filter are increased to reduce future position prediction errors when the velocity changes.

〔Example〕

FIG. 1 is a diagram showing the overall configuration of the present invention. In FIG.
+21 is installed at a certain distance, 15+,
+61 selects the transmission/reception mode using a switch. (7) is a transmitter, [81, (91Fi receiver), transmitter (
7) transmits at a predetermined frequency fB, and the receiver +81,
+91 receives reflected waves fRA and fRBt from the target of the radio waves transmitted by the transmitter (7), respectively.

αG and αυ are Doubler frequency detectors, and the Doupler frequency fDA (n) at sampling time n is determined by the transmission frequency f8 and the reception wave number fRA = fBB.
.

fDB (n) Calculated using the t-α3 formula and α3 formula, respectively.

fDA(n)=fs-fRA
("fpB(n) = fs - fRB
α3 Buddha 3 is a target position calculator, formula (1),
The polar coordinate system data RD and OD obtained from the radar are converted into orthogonal coordinate system data using equation (2). a3
is the first means for detecting the position and Doupler frequency of the moving target, and 9, a4 is the second means for detecting the Doupler frequency of the moving target. σS and αG are beam axis velocity inverse calculators, respectively:
(n) Calculate all.

In Equations 04 and 4, C is the radio wave propagation constant in air*
fDA(n) and fpB(n) are α2 formulas, (13
The Doupler frequency calculated by the formula, f8 is the transmission frequency.

aη is a target velocity vector calculator, which calculates the X-axis direction component and Y-axis direction component of the target velocity using the equations αe~(to).

The α average is a tracking filter, and the outputs of the target position calculator t13 are XD (n) and YD (n)r, and (1)
It is used to calculate equations (6) to (6).

1g” VA (n doors + 30 SoA (LeYA = V
A (n) ・aIIIOA
(lηXB −VB(n) −(X180B+1117
B -vB(n) -stn OBa'1mA -l1
11B (19 is a velocity vector determiner and a tracking filter (L)
Smooth velocity vector in sO (cross(n), :f(n))
and the speed vector (:AD(n), Ayu(n)) which is the output of the target speed vector calculator αη, and calculate the allowable error e.

If all large differences are detected, it is assumed that the moving target is performing accelerated motion, and (1) 2 to (1) of Equation 61 increases the values of coefficients α and β.

Figure 2 shows the target velocity vector calculator α.

This figure shows how to calculate the velocity vector of a moving target. In Figure 91, a is the Y axis, b is xN*pn is the position of the moving target at sampling time n, ■ is the velocity vector of the target, MA
(n) and VB (n) are respectively antenna +11,
+21 and the target Pnf on the straight line connecting the velocity vector v'fe
The projected velocity, OA, and OB are antenna (1), respectively.
The straight line connecting +21 and bau'ni is the angle made with the X axis.

FIG. 3 is a diagram specifically showing the contents of the tracking filter, in which calculations of equations (8) are performed. (31 2~(8
) in the operation of the expression,! , each of the seven coordinate components is independent.

In order to perform similar calculations, only the X component is expressed as h in FIG.

In the figure, ■ indicates input data xD(p). QD is an adder and subtracts from the previous predicted position xP(n) t-XD(n). (2), profit is the position modeling coefficient α, velocity weighting coefficient β, Q4) is the sampling period T, and @ is the divider.

Perform various calculations.

(to), @ is a multiplier, each α, //T is XD(n
)−Xp (n). (c) C31 is a delay element that holds the value of the previous sampling time. ((11 is the smoothed velocity x(n), (to) is the multiplier, (to).

(b) is an adder.

In the end, the output will be Xp (measurement).

Based on the above, calculations of equations (3)2 to (8) are performed.

Figure 4 is a diagram specifically showing the speed vector determiner ash. (n)
)and.

Each component of the speed vector (Mn (n), ÷D (!l)) calculated by the target speed vector calculator aη is input. (b) is a vector arithmetic unit which takes the outputs of the adders (to) and (to) as inputs and calculates by the Q formula.

(5 is the value ε that gives the allowable error.

(to) is a comparison operator which outputs /I'i when p and the condition of formula 311 are satisfied, and outputs a when it is not satisfied.

εD(n)>ε. 6υ (to) and (a) are each selector elements, and when the value input to A is 1, it is output to data iY on the 1 side. (IA#YI01' in the figure represents the terminal of the element) (45), (46), (47), (48)
are the input data to the selector element, and α and α are monotonically decreasing. , monotonically decreasing β, β0 and α. ,β
. is the eigenvalue of the system that satisfies the equation (to) and the equation (to).

1≧α. ≧α C3:J1≧
β. ≧β target (49),
(50) is the output of the selector element, which is a position molding coefficient α and a velocity weighting coefficient β, respectively, which are held at α and β in FIG.

(312~Weighting coefficients α, β of the tracking filter in equation (8)
When the value of IJBVc is small, the 9 constant velocity linear movement IJBVc can be followed accurately, and when it is large, it is possible to follow the acceleration movement. Therefore, when the moving target moves with acceleration with the above device, 1 is applied to the terminal A of the selector element. By increasing the values of the input zero weighting coefficients α and β, it is possible to follow the accelerated movement of the target.

〔Effect of the invention〕

As described above, the moving target tracking device of the present invention has the effect of being able to track the target without delay even when the target shifts from uniform linear motion to accelerated motion.

Although the case of the tracking algorithm based on equations (31 to (8)) has been described above, the present invention can also be applied to a tracking device having other algorithms that assume all 9 uniform linear motions.

[Brief explanation of drawings]

Fig. 1 is a diagram showing the configuration of the present invention, Fig. 2 is a diagram showing the tracking status of a moving target in a rectangular coordinate system, Fig. 3 is a diagram showing the specific configuration of the tracking filter, and Fig. 4 is a diagram showing the speed FIG. 3 is a diagram showing a specific configuration of a vector determiner. In the figure, +11 and +21 are antennas, +31 and 1
41 are first and second radars, 151 and 161 are switchers, (7) is a transmitter, +81. (9) is the receiver, αO1αυ is the Doupler frequency detector. α2 is a target position calculator, (13 is a first means for detecting the position and Doupler frequency of the moving target, and C14 is a second means for detecting the Doupler frequency of the moving target. αS, αQ are beam axis velocity calculators. , αη is the target speed vector calculator, αS is the tracking filter, +19 is the speed vector determiner, (to) is the data, an is the adder, ■, (
C) is weighting coefficient data, C is sampling period data, (C) is a divider, ＼, @ is a multiplier, (to) is an adder,
@, (7) is a delay element, 6υ is data, C (are multipliers, (g), C34 is an adder, (to), (to) is an adder,
6η is a vector calculator. (to) is a comparison operator, ((I, f4G is a selector collection, (41). (42), (45), (44), (45)
, (46), (47), (48), (
49), Criminal C). (5ri is data, a is y axis, b is I axis, v, vA
. VB is a velocity vector, and OA and OB are angles.

Claims (1)

[Claims] a first radar for observing a target moving in space; a first means for detecting the position and Doupler frequency of the moving target based on data obtained by the first radar; a second radar installed spatially independently from the radar; and second means for detecting a Doupler frequency based on data obtained by the second radar;
means for calculating the velocity vector of the moving target based on the Doupler frequency obtained by the first and second means;
and a tracking filter that assumes uniform linear motion, and is characterized in that the weighting coefficient of the tracking filter is changed by evaluating the velocity vector and the velocity vector obtained by the tracking filter. A moving target tracking device.