JPH04113291A - Multi-target homing method and device thereof - Google Patents

Abstract

PURPOSE:To intend improvement of capability of multitarget homing by outputting smoothed dimensions when the homing is established by hypothetical reliability to judge whether the homing is completed or not, and by repeating from the initial stage when not completed. CONSTITUTION:A device 10 for making a track hypothesis makes a hypothesis of a track that is whether searching data is a new target, an already homed target or unnecessary signal such as clutters. A device 11 for calculating the reliability of hypothesis calculates the reliability of hypothesis based upon the hypothesis and the searching data, and removes the unnecessary signal such as the clusters. Subsequently, smoothed values of each target are calculated by a device 12 for performing target smoothing based on the reliability of the hypothesis, the searching data and the predicted dimensions, after the unnecessary signals are removed. A device 13 for judging establishment of homing judges whether the homing is established or not, on the basis of the reliability of the hypothesis, and thereafter, when the homing is established, the smoothed dimensions are output to a device 14 for displaying the track and it is judged whether the homing is completed or not. When not completed, the procedure is repeated from the initial stage.

Description

[Detailed Description of the Invention] [Field of Industrial Application] This invention uses detection data to generate a trajectory hypothesis indicating whether the tracked target is an unnecessary signal such as clutter, and based on the reliability of the hypothesis. The present invention relates to a multi-target tracking method and device for tracking multiple targets while removing unnecessary signals such as clutter.

[Conventional technology]

In the conventional method, as shown in Fig. 3, detection data such as the target position and detection time are input (step 12), and each target is predicted based on this detection data (detection time) and the smooth specifications of each target. Calculate the specifications (step 13), convert the detection data from polar coordinates to cross coordinates (step 14), and create a software gate for each target based on the above predicted specifications (predicted existence range of the target at subsequent detection times). Calculate Step 1
6) Generate a cluster based on the detection data and the software gate (step 17).

Based on the above cluster generation results, a trajectory hypothesis is generated as to whether the above detection data is a new target, a previously tracked target, or an unnecessary signal such as clutter (step 18), and its reliability is calculated based on this hypothesis and the above detection data. At the same time, unnecessary signals such as clutter are removed (step 19), and smooth specifications of each target are calculated based on the hypothesis reliability after removing unnecessary signals such as clutter, the above-mentioned detection data, and the above-mentioned prediction specifications. In step 20), it is determined whether tracking has been established based on the reliability of the above hypothesis.In step 21), when tracking has been established.

Step 22) to output the above-mentioned smoothing specifications; step 23) to determine whether the process is finished; if not finished.

Also, K was asked to repeat from the beginning.

The conventional device includes a phased array antenna (1) for receiving radar video, as shown in FIG.

A radar signal processing device (2) that processes the radar video obtained from this 7-aided array antenna fl)
), a coordinate conversion device (41) that converts the detection data in the polar coordinate system calculated from this radar signal processing device (2) into one parent coordinates, and a coordinate conversion device (41) that converts the detection data calculated from the coordinate conversion device (4) and each A target prediction processing device rt+31 that calculates the predicted specifications of each target based on the smoothed specifications of the target, and a software gate (next and subsequent ones) of each target based on the predicted specifications calculated from this target prediction processing device A software gate calculation device that calculates the target's predicted existence range at the detection time
8), the software gate calculated by this software gate calculation device (8), and the coordinate conversion device (4).
A cluster generation device (9) that generates clusters based on the detection data calculated from 1, and a cluster HE, which is calculated by the coordinate transformation device (4) based on the cluster generation results obtained from the 1 device (9). A track hypothesis generation device αG generates a trajectory hypothesis regarding whether the detected data is a new target, a previously tracked target, or an unnecessary signal such as clutter. a hypothesis reliability calculation device lυ that calculates the degree of
The hypothesis reliability after removing unnecessary signals such as clutter calculated from the above detection data and the target prediction processing device (
3); a target smoothing processing device α2 that calculates the smoothing specifications of each target based on the predicted specifications calculated from 3); a tracking establishment determination device a3 that determines whether tracking has been established based on the hypothesis reliability; It consisted of a track display device α4 that displayed a track connected to the establishment determination device α1.

[Problem to be solved by the invention]

In the conventional multi-target tracking method and device shown in Figs. 3 and 4, the input detection data is always processed by the coordinate conversion device, so the number of input detection data is reduced. If there are many coordinates, the coordinate transformation process takes too much time and cannot be processed in real time. In particular, at the start of tracking, the predicted range of the target is not determined with certainty, so there is a high possibility that a large amount of detection data will be input. As a result, performance at the start of tracking multiple targets was extremely poor.

The present invention was made to solve this problem, and an object of the present invention is to shorten the processing time at the start of tracking and improve multi-target tracking ability.

[Means to solve the problem]

The multi-target tracking method and device according to the present invention are intended to reduce processing time at the start of tracking.

The coordinate system selection device is gold-plated.

[Effect]

In this invention, during tracking start processing, coordinate transformation is not performed for the first few movements, and processing is performed at high speed while maintaining the polar coordinate system, which is the observation specification, and after the movement becomes stable, the target Switch to a Cartesian coordinate system that allows accurate calculation of specifications.

As a result, the time required to establish tracking can be shortened and the multi-target tracking ability can be improved.

〔Example〕

FIG. 1 is a block diagram of a multi-target tracking device showing an embodiment of the present invention. In FIG. 1, (1) to (4I).

(8) to a4 are the same as shown in FIG. (5
) is a coordinate conversion device that converts the predicted specifications of each target output in the polar coordinate system from the target prediction processing device (3) into a north reference vertical or tilted coordinate system, and (6) is the radar signal processing device (2) and target prediction. The coordinate system of the detection position and predicted specifications of each target inputted from the processing device (3) is selected. Coordinate system selection device.

Reference numeral (7) is a selection data table in which data defining criteria for selection in the coordinate system selection device it+61 is stored in advance.

FIG. 2 is a diagram showing a processor Ill of an embodiment of the multi-object tracking method according to the present invention. In Fig. 2, all detection data such as position and detection time are input from the radar signal processing device (2) that processes the radar video data received from the phased array antenna +11 (step 12).
, the target prediction processing device +31 calculates the predicted specifications of each target based on this detection data (detection time) and the smooth specifications of each target (Step 13).

The first coordinate transformation device (4) and the second coordinate transformation device (
5) Convert from polar coordinate system to rectangular coordinate system. Step 14
), the coordinate system selection device (6) selects one of the two coordinate systems mentioned above (step 15), and the software gate calculation device (8) selects the software gate (next and subsequent ones) using the predicted specifications for each target. Step 16) to calculate the predicted existence range of the target at the detection time.

The cluster generation device (9) generates clusters based on the detection data, the initial software gate, and the software gate (step 17), and the track hypothesis generation device σ
Step 18) generates a trajectory hypothesis as to whether the detection data is a new target, an already tracked eye tilt, or an unnecessary signal such as clutter based on the cluster generation results in Step 18), and the hypothesis reliability calculation device αυ calculates this hypothesis and the above detection data. Step 19) calculates the reliability of the hypothesis based on the data and removes unnecessary signals such as clutter, and calculates the reliability of the hypothesis after removing unnecessary signals such as clutter with the target prediction processing device Step 20) Calculate the smooth specifications of each target based on the predicted specifications. Step 21) The tracking establishment determination device αJ determines whether tracking has been established based on the reliability of the hypothesis. Step 21) When tracking is established, Step 22) to output the smoothed specifications to the track display device α45.

Step 23) to judge whether it is finished or not; if it is not finished,
Also, repeat from the beginning.

The above processing will be specifically explained below.

The target prediction processing device (3) calculates the predicted specifications of each target using equation (1).

Here, Xkrri-1 is the predicted specification vector Φ -1 is the transition matrix Xk-1 m (+) is the smooth specification vector at time 1 km m is the tracked target number pkm (-) is the predicted specification matrix) Pk-1
m(+) is the smoothing dimension (error covariance matrix) rl(k-1)
is the process noise transformation matrix Qk-1n Process noise in the Kalman filter (component of fluctuation of target motion) T indicates the meaning of the transposed matrix.

The first coordinate transformation device (4) and the second coordinate transformation device (
The method of coordinate transformation in 5) is as follows.

Azimuth of target It is.

The basis of the judgment using the coordinate system option (lid (6)) can be determined by the number of tracking operations (switching at the third time -i).

An example of its operation is shown in Table 1.

Table 1 The software gate detection method in the software gate calculation device (8) is shown below.

Here, Zk()m is the center position vector of the software gate, Hk is the observation transformation matrix Sk, and m is the matrix 7'2(k) indicating the spread of the software gate. The device (9+) shows a cluster generation device and calculation method.

Here, zk is the detected position vector zk(-)l
is the predicted position vector Zk(-)m of the tracked target l, and the predicted position vector d of the tracked eye am is the determination threshold.

As mentioned above, the detection data zk corresponds to the tracked targets 1 and m
If both K of the software gates apply, S
k, m and Sk, ik are integrated to generate a cluster.

5′ = Sk, m + Sk, I −−
-(5) Here, Sk' is a matrix indicating the spread of the cluster.

A hypothesis generation method in the track hypothesis generation device in is shown.

(7) Identification of detected data Clutter F Also Jn Tracked target Tg New goal Nll Here, n is the number of detected data.

g is the number of targets already tracked.

h is the new target number.

Detected data is identified as one of the above three types. Let this identification result be hypothesis xk.

(a) Generation of Hypothesis A hypothesis Xi (is generated) by combining the hypothesis Xk-1 before sampling with the identification result of the latest detection data. A plurality of hypotheses are generated by combining detection data.

xk= (Xk-1HXi<)=(Xk-1)(FsT
g+Nh)...(6) Hypothesis reliability calculation method in the hypothesis reliability IFJ degree calculation device αυ is shown below.

Pk(i,j)=N(R1)kl-ROkj,BRkj
)≧KQ ・+71 where P: Confidence of hypothesis NCRpki-RoJ, 5Rkl): Time Tk
Position distribution RE""" between the i-th predicted position at -1 and the first detected position data at time Tl (Rxpkl*RYpkl+Rzpk+)
: i-th target predicted position vector ROkJ at time Tk-1 (RXOkJ TRY OkJ, R
ZOkj)', j-th target detection position data vector BRkJ at time Tl(: spread of the first S/W gate at time Tk1 N: normal distribution probability density function KO=constant When the above formula holds, tracking start Consider a hypothesis Xk, which is a combination of detection data for , as a candidate.

There is a correlation between the i-th predicted target at time tk-1 and the j-th detected data at time tk, and they are regarded as a "track."

The smoothing value calculation method in the target smoothing processing device σ3 is shown below.

(support) Kalman gain calculation for smoothing Kk, m=Pk
(-)mHkSk, m-1----- +) is the smoothed error covariance matrix.

A tracking establishment determination method in the α field of the tracking establishment determination device will be described.

(7) Check whether the same wake exists in all hypotheses.

(b) If the above conditions are met, tracking is established.

(+71 If the above conditions are not met, return to the process from the beginning.

Further, when it is determined that "one tracking has been achieved", the target smoothing specifications calculated from the target smoothing processing device ffσ are output to the track display device α◆.

〔Effect of the invention〕

As described above, according to the present invention, by processing the detection data at the time of starting tracking in the polar coordinate system, it is possible to execute the tracking start process for multiple targets at high speed, and as a result, it is possible to execute the tracking start process for multiple targets at high speed. It is possible to shorten the time and improve multi-target tracking ability.

[Brief explanation of drawings]

FIG. 1 is an overall configuration diagram of an embodiment of a multi-target tracking device according to the present invention, and FIG. 2 is a diagram showing a processing procedure of an embodiment of a multi-target tracking method according to the present invention. FIG. 3 is an overall configuration diagram of an embodiment of a conventional multi-target tracking device, and FIG. 4 is a diagram showing a processing procedure of an embodiment of a conventional multi-target tracking method. In the figure, (1) is a phased array antenna. (2) is a radar signal processing device, (31 is a target prediction processing device. (4: is a first coordinate conversion device, (5) is a second coordinate conversion device, (6) is a coordinate system selection device, +71? i selection data table, (8) is software gate calculation device, (
9) is a cluster generation device, αυ is a wake hypothesis generation device,
All is a hypothesis reliability calculation device, α is a target smoothing device, (13 is a tracking establishment determination device, and α4 is a track display device. In Figure 1, the same reference numerals indicate the same or equivalent parts. Patent applicant Prevention;, 1 piece also i;・ii':, ir;
', Ryozo Tsutsui, Director of 4 Headquarters, Figure

Claims (2)

[Claims] (1) Input detection data such as the location and detection time of targets and unnecessary signals, calculate the predicted specifications of each target based on this detection data and the smooth specifications of each target, and calculate the observed specifications of the detection data. Convert the polar coordinate system to a rectangular coordinate system, select either of the above coordinate systems, and calculate the software gate for each target (the predicted existence range of the target at the next detection time) based on the above predicted specifications. Then, based on the above detection data and the above software gate, a cluster (aggregate of predicted target existence range) is generated, and based on the generation result of the above cluster, it is determined whether the above detection data is a new target, a previously tracked target, or unnecessary signals such as clutter. Generate a hypothesis of the wake, calculate the reliability of the hypothesis based on this hypothesis and the above detection data, remove unnecessary signals such as clutter, and calculate the reliability of the hypothesis after removing unnecessary signals such as clutter, Calculating the smooth specifications of each target based on the detection data and the predicted specifications, determining whether tracking has been established based on the reliability of the hypothesis, and outputting the smooth specifications when tracking is established; A multi-target tracking method characterized in that it is determined whether or not it has ended, and if it has not ended, the above process is repeatedly performed from the beginning. (2) a phased array antenna that receives radar video, and a radar signal processing device that processes radar signals of the radar video obtained from the phased array antenna;
A target prediction processing device that calculates the predicted specifications of each target based on the detection data calculated from this radar signal processing device and the smooth specifications of each target, and a rectangular coordinate system from the polar coordinate system that is the observation specifications of the detection data. a coordinate system selection device for selecting the coordinate system, a selection data table storing data defining criteria for selection of the coordinate system selection device, and the target prediction. A software gate calculation device that calculates the software gate (predicted existence range of the target at the next detection time) of each target based on the predicted specifications calculated from the processing device, and software calculated from this software gate calculation device. a cluster generation device that generates clusters based on the detection data calculated from the gate and the radar signal processing device; and a cluster generation device that generates clusters based on detection data calculated from the gate and the radar signal processing device; and detection data calculated from the radar signal processing device based on the cluster generation results obtained from the cluster generation device A track hypothesis generation device that generates a hypothesis of a track such as whether it is a new target, a previously tracked target, or an unnecessary signal such as a clutter, and calculates the reliability of the hypothesis based on the hypothesis obtained from this track hypothesis generation device and the above detection data. A hypothesis reliability calculation device that removes unnecessary signals such as clutter, and a hypothesis reliability calculated from this hypothesis reliability calculation device after removing unnecessary signals such as clutter, calculated from the above-mentioned detection data and the above-mentioned target prediction processing device. a target smoothing processing device that calculates the smoothing specifications of each target based on the predicted specifications, a tracking establishment determination device that determines whether tracking has been established based on the above-mentioned hypothesis reliability, and a wake connected to this tracking establishment determination device. A multi-target tracking device characterized by comprising a track display device that displays.