JP4166651B2 - Target tracking device - Google Patents

Description

  The present invention relates to information calculation of distance change rate (Doppler velocity) performed as preprocessing in an apparatus for estimating a target locus from detection data obtained by observation with a pulse compression radar, and in particular, unnecessary signals are simultaneously observed. The present invention relates to a target tracking device that can accurately and efficiently process a correlation determination made up of a combination of existing track and detection data in a situation where a plurality of targets are mixed.

In general, in a target tracking device that estimates a track that is a target trajectory using detection data obtained from a pulse compression radar (hereinafter also simply referred to as “radar”), which detection data is used to form a track The tracking performance greatly depends on the determination method. One of the following three interpretations (1) to (3) applies to each detection data.
(1) Correlate with any existing track.
(2) A newly generated target.
(3) Unnecessary signal.
Determining which interpretation is correct among the above interpretations (1) to (3) is called “correlation determination”. In the correlation determination, each existing track is used as a determination index. A numerical value indicating the degree of correlation for the combination with the detection data is required.
Often used as an index representing the degree of correlation is the closeness between the predicted position of the existing track and the distance of the detection data.

A conventional target tracking device includes a sensor that generates detection data, an individual correlation determination unit that includes a gate internal / external determination unit and a likelihood calculation unit for processing the detection data, and wake-related data related to the individual correlation determination unit, A track specification calculation unit and a tracking track determination unit, and, for example, update the existing track with appropriate detection data according to the interpretations (1) to (3), and detect detection data that does not correspond to the existing track. The combination of correlations that are interpreted as new targets or unwanted signals are determined to be valid.
That is, the sampling data reading step, the individual correlation determination step, and the tracking track determination step are set as a processing process (the above-described procedure processing) for one sample of the detection data group obtained by observation at the same time by the radar. The processing loop is repeatedly executed for each observation time (sample).

は平滑値、

は平滑誤差共分散行列、

はカルマンゲイン、

は残差共分散行列である。
なお、既存航跡の運動諸元を推定する際の座標の取り方については、状態変数を、直交座標による位置ｘ，ｙ，ｚ、および、速度の各成分ｖｘ，ｖｙ，ｖｚとする。
このとき、平滑値、平滑誤差共分散行列、カルマンゲインおよび残差共分散行列は、以下の式（１）〜（４）により算出される。 Hereinafter, the “individual correlation determination step” by the conventional target tracking device will be described.
The wake specification calculation unit updates the motion specification of the existing wake using the position information of the detection data correlated with the previous sample in the processing of the current previous sample. That is, the smoothed value at the latest time and the predicted value at the next time are calculated.
The smooth value and the predicted value are calculated by the following formulas (1) to (6).
First, in the smooth value calculation,

Is the smooth value,

Is the smoothing error covariance matrix,

Is Kalman gain,

Is the residual covariance matrix.
In addition, about the method of taking the coordinate at the time of estimating the motion specification of the existing wake, a state variable shall be the position x, y, z by a rectangular coordinate, and each component vx, vy, vz of velocity.
At this time, the smooth value, the smooth error covariance matrix, the Kalman gain, and the residual covariance matrix are calculated by the following equations (1) to (4).

は観測行列、

は観測誤差共分散行列である。
また、

は予測値、予測誤差共分散行列であり、また、

は、極座標で示された共分散行列を直交座標用に変換を行う行列である。
予測値および予測誤差共分散行列は、予測値計算用の以下の式（５）、（６）により算出される。 here,

Is the observation matrix,

Is the observation error covariance matrix.
Also,

Is the predicted value, the prediction error covariance matrix, and

Is a matrix that transforms the covariance matrix indicated by polar coordinates into Cartesian coordinates.
The prediction value and the prediction error covariance matrix are calculated by the following equations (5) and (6) for calculating the prediction value.

は推移行列、

は駆動雑音共分散行列である。 here,

Is the transition matrix,

Is the driving noise covariance matrix.

がこのゲート条件を満たせばゲート内、満たさなければゲート外と判定する。
なお、ゲート内外判定用の条件は、以下の式（７）で表される。 Hereinafter, the determination step and the correlation likelihood calculation step by the gate internal / external determination unit and the correlation likelihood calculation unit are executed for a combination of the existing track and all of the detection data.
First, the gate inside / outside determination unit sets a gate, which is a region where detection data is highly likely to be observed, from the predicted value of the existing track and the residual covariance matrix, and the position vector of the detection data

If this gate condition is satisfied, it is determined that it is inside the gate, and otherwise, it is determined that it is outside the gate.
The condition for determining whether the gate is inside or outside is expressed by the following equation (7).

は事前に設定されるパラメータである。また、

は、予測位置であり、以下の式（８）により算出される。 here,

Is a parameter set in advance. Also,

Is a predicted position and is calculated by the following equation (8).

  When it is determined that the detection data is out of the gate, the combination of the detection data and the existing track is regarded as “no possibility of correlation”, and is excluded from consideration in the subsequent correlation determination.

Next, correlation likelihood calculation will be described.
When it is determined that the detection data is within the gate, the combination of the detection data and the existing track is “possibly correlated”.
At this time, the likelihood calculating unit (correlation degree calculating unit) calculates the degree of correlation in a combination having a correlation possibility by the following equation (9).

  This calculation method assumes that the position of the detection data is a statistic according to the Gaussian distribution of the covariance matrix Sk centered on the predicted position, and the “correlation degree (likelihood)” that is the calculation result is It shows how close the position of the detection data is to the predicted value.

Finally, the tracking track determination unit selects a combination of the existing track and the detection data using the combination of the existing track with the possibility of correlation and the detection data and the correlation degree calculated for the combination, Determine the overall correlation.
In this correlation determination, not only the correlation between the existing track and the detection data but also the situation where the detection data cannot be obtained from the existing track and the possibility that the detection data is a new target or an unnecessary signal are considered.

There are various methods for determining the overall correlation, but one example is a method “Nearest Neighbor” that allocates detection data having the highest degree of correlation to each existing track (for example, Non-patent document 1).
Further, as another example, a method “MHT (MHT ()” that considers all possible correlations for each existing track, combines the correlations to form a hypothesis, and calculates the reliability of the hypothesis based on each correlation degree. (Multiple Hyperthesis Tracking) ”(see, for example, Non-Patent Document 2).

As described above, the conventional target tracking device estimates the predicted position to be detected next in order to determine whether or not there is a correlation between the existing track and the detection data in the individual correlation determination step. As shown in equations (5) and (6), the estimated value (smooth position) of the target position and the estimated value (smooth speed) of the target in the previous sample are used.
That is, the position that has moved from the smooth position according to the smooth speed for the time from the previous observation time to the current observation time is set as the predicted position.

As is well known, in addition to position information, a Doppler frequency may be obtained from a radar, and a Doppler velocity is calculated from the Doppler frequency. The Doppler speed obtained by this method is referred to as “Doppler speed by Doppler frequency”.
The Doppler speed is the rate of change of the distance between the target and the radar and includes information about the speed. Therefore, it has been proposed to improve the tracking performance by utilizing this information.

の積の分だけずれるので、互いに異なる値となる。
すなわち、アップチャープ時に観測される距離は、以下の式（１０）で表される。 The Doppler speed can also be calculated by another method. For example, the linear frequency modulation method is a method in which up-chirping and down-chirping are repeated in a short time and the obtained observation position shift is used.
The distance observed during up-chirp and the distance observed during down-chirp are the distance change rate and the chirp constant.

Therefore, the values are different from each other.
That is, the distance observed at the time of up-chirp is expressed by the following formula (10).

  Further, the distance observed at the time of down-chirp is expressed by the following equation (11).

  Therefore, by dividing the distance deviation between the two observation distances by a constant, the Doppler velocity can be obtained as shown in the following equation (12).

The Doppler speed obtained by this method will be referred to as “Doppler speed due to Doppler shift difference”.
The “Doppler speed due to the Doppler shift difference” has advantages and disadvantages, respectively, similar to the “Doppler speed due to the Doppler frequency” described above.
In other words, the Doppler velocity based on the Doppler frequency has the advantage of high observation accuracy, but when the pulse repetition frequency (PRF) of the radar is low, the high frequency component of the received wave is folded, so that the target to be observed is Even if the speed is very fast, it will be observed as a Doppler speed comparable to that of a slow target, so there are multiple candidate speeds for one observed Doppler speed, making it difficult to identify There is.
In addition, since the Doppler speed due to the Doppler shift difference does not cause a problem due to folding, the speed is uniquely determined with respect to the observation value of one Doppler speed, but the observation accuracy is lower than the Doppler speed due to the Doppler frequency. .

On the other hand, for the Doppler speed obtained from the detection data, the distance change rate (predicted speed) is calculated separately from the predicted value of the distance change rate track, the similarity between these two speed information is calculated, and the degree of correlation is calculated. There has also been proposed a target tracking device used for the above (see, for example, Patent Document 1).
However, this tracking technique does not suggest a means for obtaining a high-quality Doppler velocity observation value utilizing the characteristics of the velocity information as a method for obtaining the Doppler velocity information.

Japanese Patent Publication No. 7-101233 Samuel S. Chapter 4 Section 3 of "Multiple-Target Tracking with Radar Application" by ARTMAN (ARTTECH HOUSE) Donald B.D. Reid's paper “An Algorithm for Tracking Multiple Targets” (IEEE Transactions on Automatic Control, Vol. AC-24, No6, December, 1979)

In the conventional target tracking device, when the Doppler speed based on the Doppler frequency is used, even if the speed of the target to be observed is very high, the Doppler speed is about the same as that of the slow target. There is a problem that there are a plurality of candidate velocities for the observed value of the speed, making it difficult to specify.
In addition, when Doppler velocity due to Doppler shift difference is used, there is a problem that high observation accuracy cannot be obtained.
Further, as in Patent Document 1, when the similarity between the Doppler velocity obtained from the detection data and the predicted velocity is used for the correlation degree calculation, a means for obtaining a good observation value of the Doppler velocity is provided. As a result, there was a problem that high observation accuracy could not be obtained.

  This invention is a tracking method using the Doppler velocity, and makes use of the features of the Doppler velocity due to the Doppler frequency and the Doppler velocity difference due to the Doppler shift difference to detect the quality Doppler velocity. The object is to obtain a target tracking device capable of more accurately determining the correlation between data and an existing track.

  A target tracking device according to the present invention obtains position information and Doppler velocity from detection data obtained by observation with a pulse compression radar, and tracks the target using the observed value of position information and the observed value of Doppler velocity. A Doppler speed information extraction unit that narrows down first Doppler speed candidates from detection data, a gate inside / outside determination unit that determines correlation of detection data with respect to an existing track from the detection data and the first Doppler speed, A tracking wake calculation unit that performs calculation for determining the tracking track of the target using the wake and detection data, and the Doppler velocity information extraction unit calculates the first from the frequency deviation between the transmission frequency and the reception frequency of the pulse compression radar. Of Doppler velocity for the distance observation value obtained by alternately performing up-chirp and down-chirp The second Doppler speed is calculated from the deviation, the first and second Doppler speeds are compared to narrow down the first Doppler speed candidates, and the gate inside / outside determination unit extracts the predicted speed and Doppler speed information of the existing wake The first Doppler speed candidates obtained from the first section are compared with each other, the first Doppler speed candidates are uniquely narrowed down, and the predicted speed and the first Doppler speed candidate are narrowed down uniquely. Determine whether the detection data can be correlated with the existing track from the deviation, calculate the possible area of the detection data from the predicted position of the existing track, and whether the detection data can be correlated with the existing track This is a judgment.

  According to the present invention, by utilizing the characteristics of the Doppler speed due to the Doppler frequency and the Doppler speed due to the Doppler shift difference, the present invention can be used to correlate the detection data with the existing track by devising a good Doppler speed observation value. The determination can be made more accurately.

Embodiment 1 FIG.
Embodiment 1 of the present invention will be described below.
FIG. 1 is a block diagram showing a first embodiment of the present invention, FIG. 2 is a flowchart showing a schematic procedure of tracking processing according to the first embodiment of the present invention, and FIG. 3 is an individual flowchart according to the first embodiment of the present invention. It is a flowchart which shows a correlation determination process.
FIG. 15 is an explanatory diagram showing specific correlation processing according to the above-described three interpretations (1) to (3). Two existing wakes T1 and T2 exist and two intersecting gates (dotted lines) 3), the three detection data 00, 10, and 11 are obtained.

In FIG. 15, the detection data 00 is closest to the predicted value of the existing wake T1 (extension portion on the wake tip side). For the existing wake T1, the combination with the detection data 00 is “the highest degree of correlation. "It is shown that.
Further, the detection data 11 is closest to the predicted value of the existing track T2, and the combination with the detection data 11 indicates that the degree of correlation is the highest for the existing track T2.

Here, as the detection data to which the existing tracks T1 and T2 correlate, according to the policy of “determining the combination of the existing track and the detection data with the highest degree of correlation”, the following (1a) to (3a) ), It is possible to determine that the combination of correlations is valid.
(1a) The existing track T1 is updated using the detection data 00.
(2a) The existing wake T2 is updated using the detection data 11.
(3a) The detection data 01 is interpreted as a new target or an unnecessary signal.

In FIG. 1, the target tracking device according to the present invention includes a sensor 1, a Doppler velocity information extraction unit 2, an individual correlation determination unit 3, a wake related data 4, a wake specification calculation unit 5, and a tracking wake determination unit 6. The individual correlation determination unit 3 includes a gate inside / outside determination unit 31 and a likelihood calculation unit 32.
In this case, it can be considered that the Doppler velocity information extraction unit 2 is added as compared with the conventional device.

The Doppler velocity information extraction unit 2 is related to the sensor 1 and the gate inside / outside determination unit 31 in the individual correlation determination unit 3.
The gate inside / outside determination unit 31 is related to the likelihood calculation unit 32, the sensor 1, the Doppler speed information extraction unit 2, the track related data 4, and the tracking track determination unit 6.
Further, the wake related data 4 is related to the gate inside / outside determination unit 31 and the wake specification calculation unit 5, and the tracking wake determination unit 6 performs the likelihood calculation unit 32 and the wake specification calculation in the individual correlation determination unit 3. Part 5 is related.
The likelihood calculation unit 32, the track specification calculation unit 5, and the tracking track determination unit 6 constitute a tracking track calculation unit that performs calculations for determining the target tracking track using the existing track and the detection data.

The sensor 1 is related to a pulse compression radar (not shown), and generates detection data obtained by observation by the radar.
The detection data from the sensor 1 includes position information, frequency information, and a distance observation value. The position information is input to the gate inside / outside determination unit 31 in the individual correlation determination unit 3, and the frequency information and the distance observation value are the Doppler velocity. Input to the information extraction unit 2.

  The Doppler velocity information extraction unit 2 narrows down and outputs first Doppler velocity candidates from the detection data (frequency information and distance observation value). Specifically, the Doppler velocity information extraction unit 2 obtains a first Doppler velocity candidate from the frequency deviation between the transmission frequency and the reception frequency of the pulse compression radar, and obtains the Doppler velocity information extraction unit 2 by alternately executing up-chirp and down-chirp. The second Doppler velocity is calculated from the distance deviation of the obtained distance observation values, the first and second Doppler velocities are compared to narrow down the first Doppler velocity candidates, and the inside and outside of the gate is detected as detection data comprising Doppler velocity information. Input to the determination unit 31.

  The gate inside / outside determination unit 31 determines the correlation of the detection data with respect to the existing track from the detection data (position information) and the first Doppler speed. Specifically, the gate inside / outside determination unit 31 compares the predicted speed of the existing track with a plurality of candidates (here, all candidates) of the first Doppler speed obtained from the Doppler speed information extraction unit 2. The first Doppler speed candidates are uniquely narrowed down, and it is determined whether or not the detection data and the existing wake can be correlated from the speed deviation between the predicted speed and the uniquely narrowed first Doppler speed candidates. At the same time, an area where detection data can exist is calculated from the predicted position of the existing track, and it is determined whether or not the detection data and the existing track can be correlated.

The likelihood calculating unit 32 calculates the degree of correlation between the existing track and the detected data based on the correlation determination result by the gate inside / outside determining unit 31 and inputs the calculated degree to the tracking track determining unit 6.
The wake-related data 4 takes in the detection data from the gate inside / outside determination unit 31 and the smooth value and the prediction value of the wake from the wake specification calculation unit 5, inputs the prediction value to the gate inside / outside determination unit 31, and also detects the detection data. Is input to the wake specification calculation unit 5.
The wake specification calculation unit 5 calculates the motion specification when the existing wake is updated using the detection data. The wake item calculation unit 5 includes a function as a position information filter.

The tracking track determination unit 6 determines a target tracking track based on the degree of correlation between the existing track and the detection data. In other words, the tracking track determination unit 6 includes the combination information of the existing track with the possibility of correlation obtained from the gate inside / outside determination unit 31 and the detection data, the degree of correlation information obtained from the likelihood calculation unit 32, and the track The tracking track of the target is determined using the wake specification obtained from the specification calculation unit 5.
The target tracking apparatus according to the first embodiment of the present invention shown in FIG. 1 obtains position information and Doppler velocity from detection data obtained by observation with a pulse compression radar, and obtains position information observation values and Doppler velocity. The target is tracked using the observed values.

In FIG. 2, the tracking process according to the first embodiment of the present invention includes a process of reading detection data to be sampled next (step S21), an extraction process of Doppler speed information (step S22), an existing track and detection data. It consists of a determination process of individual correlation with all (step S23) and a determination process of tracking track (step S24).
Steps S21 to S24 correspond to the processes of the sensor 1, the Doppler velocity information extraction unit 2, the individual correlation determination unit 3 and the tracking track determination unit 6 in FIG.
In this case, it can be considered that Doppler velocity information extraction processing (step S22) is added to the processing procedure of the conventional apparatus.

First, specific contents of the Doppler velocity information extraction process (step S22) will be described.
In step S22, Doppler speed candidates are extracted by the following procedure. Here, it is assumed that observation is performed by repeating up-chirp and down-chirp.
First, “the Doppler speed by the Doppler frequency” is calculated. At this time, an infinite number of possible Doppler speeds can be obtained at a constant speed interval.
Next, calculate the “Doppler speed due to the Doppler shift difference”, select several values close to the “Doppler speed due to the Doppler shift difference” among the “Doppler speed due to the Doppler frequency difference”, and set these values as the “Doppler frequency difference”. Candidate ".
As a selection criterion at this time, any of the following (2-1) and (2-2) may be used.
(2-1) A certain number is selected in the order close to “Doppler speed by Doppler shift difference”.
(2-2) A value is selected such that the speed deviation from the “Doppler speed due to the Doppler shift difference” is not more than a certain threshold value.

Steps S21, S23, and S24 in FIG. 2 are known processes executed also in the conventional apparatus, but the detailed contents of the individual correlation determination process (step S23) are different from those in the conventional apparatus. However, the content of the tracking track determination process (step S24) is the same as that of the conventional apparatus.
The flowchart of FIG. 3 shows the detailed procedure of the individual correlation determination process (step S23) for the combination of the existing track and all of the detection data.
The individual correlation determination process according to the first embodiment of the present invention will be described below with reference to FIG.

In FIG. 3, the individual correlation determination processing includes Doppler speed narrowing processing (step S31), distance change rate (Doppler speed) matching determination processing (step S32), and Doppler speed candidate rejection processing (step S33). , A gate inside / outside determination process (step S34), a correlation likelihood calculation process (step S35), and a Doppler speed candidate selection process (step S36) are included, and the process returns from step S36 to step S31 and is repeated. A loop is configured.
Steps S31 to S34 correspond to the processing of the gate inside / outside determination unit 31 in FIG. 1, and steps S35 and S36 correspond to the processing of the likelihood calculation unit 32 in FIG.

First, the Doppler speed narrowing process (step S31) will be described.
In step S31, candidates for Doppler speeds calculated in the Doppler speed information extraction process (step S22) are narrowed down.
Specifically, the narrowing-down process is performed by calculating a distance change rate from the predicted value of the wake and selecting a candidate closest to this value.
The distance change rate is calculated from the predicted value of the wake as shown in the following equation (13).

  However, a predicted value shall be represented like the following formula | equation (14).

と、上記式（１３）が一致するか否かを判定するため、両者の差による評価関数を、以下の式（１５）のように設定する。 Next, in the distance change rate coincidence determination process (step S32), the Doppler speed uniquely selected in the previous narrowing process (step S31) and the distance change rate (Doppler speed) obtained from the predicted value of the wake are obtained. It is determined whether or not they match.
At this time, the degree of coincidence between the uniquely selected Doppler velocity and the distance change rate obtained from the predicted value is statistically determined using the covariance of both errors.
Specifically, the statistical calculation method using covariance is as follows.
First, the Doppler velocity candidates obtained from the detection data

In order to determine whether or not the above equation (13) matches, an evaluation function based on the difference between the two is set as in the following equation (15).

  In this evaluation function, seven variables are set as arguments as in the following Expression (16).

に等しいならば、以下の式（１７）が成立する。 Subsequently, a hypothesis is established that the evaluation function is “0”, that is, “the distance change rate calculated from the Doppler velocity candidate and the predicted value of the position information filter is equal”. This hypothesis is correct and the argument is true

Is equal to the following expression (17).

として、位置情報フィルタの予測値と、ドップラ速度の観測値とを代入する。
この場合、予測値には、フィルタの推定による誤差が含まれているので、評価関数は「０」にはならない。
しかし、上記仮説が正しければ、以下の式（１８）が成立する。 However, since the true value is not known, the argument

As described above, the predicted value of the position information filter and the observed value of the Doppler velocity are substituted.
In this case, since the prediction value includes an error due to filter estimation, the evaluation function does not become “0”.
However, if the above hypothesis is correct, the following equation (18) is established.

  Therefore, the calculation result of the evaluation function should be within the range in which the error of each component affects. However, it is assumed that the following equation (19) holds.

の関数である。関数Ｌとしては、真値

における値を設定することが理想的であるが、真値は得られないので、位置情報フィルタの予測値とドップラ速度の観測値とを代入して近似する。
以上より、評価関数の値は、平均「０」、共分散

の正規分布となる。ただし、共分散行列

は、位置情報フィルタの予測誤差共分散行列（上記式（６）参照）と、ドップラ速度の観測誤差

を用いて、以下の式（２０）のように算出される。 Where L is

Is a function of Function L is true value

It is ideal to set a value at, but since a true value cannot be obtained, the predicted value of the position information filter and the observed value of the Doppler velocity are substituted and approximated.
From the above, the value of the evaluation function is the average “0”, covariance

Of normal distribution. However, the covariance matrix

Is the prediction error covariance matrix of the position information filter (see equation (6) above) and the Doppler velocity observation error.

Is calculated as in the following equation (20).

  From equation (20), the statistic is a chi-square distribution with “1” degrees of freedom as in equation (21) below.

を満たす場合には棄却される。すなわち、位置情報フィルタの出力とドップラ速度の観測値とは一致しておらず、既存航跡と探知データとの間に相関は存在しないものと見なす。
一方、

を満たす場合には、上記仮説が採択される。すなわち、位置情報フィルタの出力とドップラ速度の観測値とは一致しており、既存航跡と探知データとの間に相関が存在するものと見なす。
ここで、

は、判定のための閾値である。 From the above, the hypothesis that the evaluation function is “0”, that is, “the predicted value of the position information filter matches the observed value of the Doppler velocity” is

If it meets, it will be rejected. That is, the output of the position information filter and the observed value of the Doppler velocity do not match, and it is assumed that there is no correlation between the existing track and the detection data.
on the other hand,

If this is the case, the above hypothesis is adopted. In other words, the output of the position information filter and the observed value of the Doppler velocity coincide with each other, and it is assumed that there is a correlation between the existing track and the detection data.
here,

Is a threshold for determination.

As described above, in the distance change rate match determination process (step S32), when it is determined that the Doppler speed of the detection data does not match the distance change rate of the existing track (that is, NO), the detection data and Assuming that there is no correlation with the track, the process proceeds to a rejection process (step S33), and the combination of the existing track and the detection data is rejected.
In step S32, if it is determined that the Doppler speed of the detection data matches the distance change rate of the existing track (that is, YES), the gate information is further determined using the position information (step S34). Execute.

  In step S34, if it is determined that the position information is not within the gate (ie, NO), the process proceeds to a rejection process (step S33). If it is determined that the position information is within the gate (ie, YES), the correlation is performed. The likelihood calculation process (step S35) is executed, the process proceeds to the adoption process (step S36), and the process routine of FIG.

  As described above, according to the first embodiment of the present invention, by comparing the “Doppler speed by the Doppler frequency” and the “Doppler speed by the Doppler shift difference”, a high-quality Doppler speed is calculated and used for tracking. Thus, the correlation between the detection data and the existing wake can be accurately determined.

Embodiment 2. FIG.
In the first embodiment, the tracking track is determined by setting one hypothesis, but a plurality of hypotheses may be set.
A second embodiment of the present invention using a tracking track determination unit based on multiple hypotheses will be described below.
4 is a block diagram showing a second embodiment of the present invention, FIG. 5 is a flowchart showing a tracking process procedure according to the second embodiment of the present invention, and FIG. 6 shows an individual correlation determination process according to the second embodiment of the present invention. The same components as those described above (see FIGS. 1 to 3) are denoted by the same reference numerals as those described above and will not be described in detail.

FIG. 4 differs from the above (FIG. 1) only in that the tracking track determination unit 6A based on multiple hypotheses is used.
FIG. 5 differs from the above (FIG. 2) only in that a Doppler speed narrowing process (step S52) is added after step S22.
6 is different from the above (FIG. 3) only in that the Doppler speed narrowing process (step S31) is deleted.
In this case, the above-described MHT (Multiple Hyperthesis Tracking) is used in the overall correlation determination in the tracking track determination process (step S24). That is, the tracking track of the target is determined while maintaining the target hypothesis.

In FIG. 5, the specific content of the Doppler speed narrowing process (step S52) is as follows.
In step S52, candidates for Doppler speeds calculated in the Doppler speed information extraction process (step S22) are narrowed as follows.
First, the distance change rate (Doppler speed) is calculated from the predicted values of all existing tracks calculated by the MHT method and referenced in a plurality of hypotheses, and can be correlated by comparison with the Doppler speed from the detection data. Narrow down existing tracks.
Next, the existing track with the highest reliability is selected from the narrowed existing tracks.
Subsequently, the candidate of the Doppler speed is uniquely narrowed down by selecting the candidate of the value closest to the distance change rate of the existing track with the highest reliability among the Doppler speed candidates.
The flowchart in FIG. 6 shows an individual correlation determination process including steps S32 to S36 similar to the above (see FIG. 3).

  As described above, according to the second embodiment of the present invention, a high-quality Doppler speed can be calculated by comparing the “Doppler speed based on the Doppler frequency” and the “Doppler speed based on the Doppler shift difference”. Further, by comparing the MHT (determining the tracking track of the target while maintaining the target hypothesis) with the track of the Doppler speed, the correlation determination between the detection data and the existing track is performed by narrowing down and using the Doppler speed candidates. Can be done accurately.

Embodiment 3 FIG.
In the first and second embodiments, the predicted value is input from the wake-related data 4 to the gate internal / external determination unit 31 and the gate internal / external determination process is performed using the predicted value. However, the gate internal / external determination is performed using the smooth value. Processing may be executed.
The third embodiment of the present invention in which the gate inside / outside determination process is executed using the smooth value will be described below.
FIG. 7 is a block diagram showing the third embodiment of the present invention, and FIG. 8 is a flowchart showing the individual correlation determination process according to the third embodiment of the present invention, which is the same as described above (see FIGS. 1 and 3). For the above, the same reference numerals as those described above are attached, and detailed description thereof is omitted.

7 is different from the above (FIG. 1) only in that it includes a gate inside / outside determination unit 31B using a smooth value and wake-related data 4B for inputting a smooth value to the gate inside / outside determination unit 31B.
FIG. 8 differs from the above (FIG. 3) only in that a wake smooth value calculation process (step S81) is added before step S31B.
The tracking processing procedure according to the third embodiment of the present invention is as shown in FIG.
Also in this case, it is assumed that the aforementioned MHT is used in the overall correlation determination in the tracking track determination process.

In the wake smooth value calculation process (step S81) in FIG. 8, smooth calculation of the existing wake is performed based on the existing wake and the position information of the detection data.
Subsequently, in the Doppler speed narrowing process (Step S31B), the Doppler speed candidates calculated in the Doppler speed information extraction process (Step S22 in FIG. 2) are narrowed down.
Specifically, the narrowing-down process is performed by calculating a distance change rate from the smooth value of the wake and selecting a candidate closest to this value.
That is, the distance change rate is calculated from the smooth value of the wake like the following formula (22).

  However, the following formula | equation (23) shall be materialized.

と、式（２２）から求まる距離変化率とが一致するか否かを判定するために、両者の速度偏差による評価関数を、以下の式（２４）のように設定する。 Next, in the distance change rate coincidence determination process (step S32B), the degree of coincidence between the Doppler speed uniquely selected in the previous narrowing process (step S31B) and the distance change rate obtained from the smooth value of the wake is determined. And statistically calculated using the covariance of both errors.
The calculation process at this time is as follows.
First, the Doppler velocity candidates obtained from the detection data

In order to determine whether or not the distance change rate obtained from Expression (22) matches, an evaluation function based on the speed deviation between the two is set as shown in Expression (24) below.

  In this evaluation function, seven variables are set as arguments as shown in the following equation (25).

に等しいならば、以下の式（２６）が成立する。 Here, a hypothesis is established that the evaluation function is “0”, that is, the distance change rate calculated from the Doppler velocity candidate and the smooth value of the position information filter is equal. This hypothesis is correct and the argument is true

Is equal to the following equation (26).

として、位置情報フィルタの平滑値と、ドップラ速度の観測値とを代入する。
この場合、予測値にはフィルタの推定による誤差が含まれているので、評価関数は「０」にはならない。しかし、仮説が正しければ、以下の式（２７）が成立する。 However, since the true value is not known, the argument

The smooth value of the position information filter and the observed value of the Doppler velocity are substituted.
In this case, since the prediction value includes an error due to filter estimation, the evaluation function does not become “0”. However, if the hypothesis is correct, the following equation (27) holds.

ここで、Ｌは

の関数である。関数Ｌとしては、真値

における値を設定することが理想的であるが、真値は得られないので、位置情報フィルタの平滑値とドップラ速度の観測値とを代入して近似する。以上より、評価関数の値は、平均「０」、共分散

の正規分布となる。ただし、共分散行列

は、位置情報フィルタの平滑誤差共分散行列（前述の式（２）参照）と、距離変化率の観測誤差

とを用いて、以下の式（２９）のように算出される。 Therefore, the calculation result of the evaluation function should be within the range in which the error of each component affects. However, it is assumed that the following formula (28) holds.

Where L is

Is a function of Function L is true value

It is ideal to set a value at, but since a true value is not obtained, approximation is performed by substituting the smooth value of the position information filter and the observed value of the Doppler velocity. From the above, the value of the evaluation function is the average “0”, covariance

Of normal distribution. However, the covariance matrix

Is the smoothing error covariance matrix of the location information filter (see Equation (2) above) and the distance change rate observation error

And is calculated as in the following equation (29).

  From equation (29), the statistic is a chi-square distribution with “1” degrees of freedom, as in equation (30) below.

を満たす場合には棄却される。すなわち、位置情報フィルタの出力とドップラ速度の観測値とは一致しておらず、既存航跡と探知データとの間に相関は存在しないものと見なす。
一方、

を満たす場合には、上記仮説を採択する。すなわち、位置情報フィルタの出力とドップラ速度の観測値とは一致しており、既存航跡と探知データとの間に相関が存在するものと見なす。
ここで、

は、一致判定のための閾値である。 From the above, the hypothesis that the evaluation function is “0”, that is, “the smooth value of the position information filter and the observed value of the Doppler velocity match” is

If it meets, it will be rejected. That is, the output of the position information filter and the observed value of the Doppler velocity do not match, and it is assumed that there is no correlation between the existing track and the detection data.
on the other hand,

If the above is satisfied, the above hypothesis is adopted. In other words, the output of the position information filter and the observed value of the Doppler velocity coincide with each other, and it is assumed that there is a correlation between the existing track and the detection data.
here,

Is a threshold for matching determination.

Thus, in the distance change rate coincidence determination process (step S32B), when it is determined that the Doppler speed of the detection data does not match the distance change rate of the existing track (that is, NO), the detection data and the track are detected. Is considered to have no correlation, and the process proceeds to a rejection process (step S33), and the combination of the existing track and the detection data is rejected.
If it is determined in step S32B that the Doppler speed of the detection data matches the distance change rate of the existing track (that is, YES), the same steps S34 to S36 as described above are further executed.

  As described above, according to the third embodiment of the present invention, by comparing the “Doppler speed based on the Doppler frequency” and the “Doppler speed based on the Doppler shift difference”, a high-quality Doppler speed is calculated and used for tracking. Furthermore, the correlation determination between the detection data and the existing track can be accurately performed by narrowing down the Doppler speed candidates compared with the smooth value of the existing track.

Embodiment 4 FIG.
In the third embodiment, the tracking track is determined by setting one hypothesis, but a plurality of hypotheses may be set as in the second embodiment.
Hereinafter, a fourth embodiment of the present invention using a tracking track determination unit based on multiple hypotheses will be described.
FIG. 9 is a block diagram showing a fourth embodiment of the present invention. The same parts as those described above (see FIGS. 4 and 7) are denoted by the same reference numerals as those described above, and detailed description thereof is omitted.
FIG. 10 is a flowchart showing the tracking processing procedure according to the fourth embodiment of the present invention. The same components as those described above (see FIGS. 2 and 5) are denoted by the same reference numerals as those described above, and detailed description thereof is omitted. To do.

9 is different from the above (FIG. 7) only in that the tracking track determination unit 6A based on a plurality of hypotheses is used.
10 is different from the above (FIG. 5) only in that a wake smooth value calculation process (step S102) is added after step S22.
The individual correlation determination process according to the fourth embodiment of the present invention is as shown in FIG.
Also in this case, it is assumed that the aforementioned MHT is used in the overall correlation determination in the tracking track determination process.

In FIG. 10, the Doppler speed narrowing process (step S52) includes a wake smooth value calculation process (step S102) as a pre-process.
In step S102, smooth values obtained by updating using detection data are calculated for all wakes calculated by the MHT method and referred to in a plurality of hypotheses.
Subsequently, in the Doppler speed narrowing process (Step S52), the Doppler speed candidates calculated in the Doppler speed information extraction process (Step S22) are narrowed down.

The narrowing-down process in step S52 is performed as follows.
First, the distance change rate is calculated from the track smooth value calculated in the previous step S102, and an existing track that can be correlated is selected by comparing this value with the Doppler velocity.
Next, the existing track with the highest reliability is selected uniquely from the selected tracks.
The candidates are narrowed down by selecting a candidate with a value closest to the distance change rate of the wake having the highest reliability among the Doppler speed candidates.

  As described above, according to the fourth embodiment of the present invention, a high-quality Doppler speed can be calculated by comparing the “Doppler speed based on the Doppler frequency” and the “Doppler speed based on the Doppler shift difference”. Further, by narrowing down the candidates by comparison with the smooth value of the MHT track and using it for tracking, the correlation between the detection data and the existing track can be accurately determined.

Embodiment 5. FIG.
Although not particularly mentioned in the first to fourth embodiments, a Doppler filter calculation unit related to the wake-related data may be added.
Hereinafter, a fifth embodiment of the present invention in which a Doppler filter calculation unit is added will be described.
FIG. 11 is a block diagram showing a fifth embodiment of the present invention. The same components as those described above (see FIG. 1) are denoted by the same reference numerals as those described above, and detailed description thereof is omitted.
FIG. 12 is a flowchart showing the tracking processing procedure according to the fifth embodiment of the present invention. The same components as those described above (see FIG. 2) are denoted by the same reference numerals and detailed description thereof is omitted.

FIG. 11 differs from the above (FIG. 1) only in that a Doppler filter calculation unit 7 is added in relation to the track related data 4C.
12 is different from the above (FIG. 2) only in that a Doppler filter calculation process (step S125) is added after step S24.
The individual correlation determination process according to the fifth embodiment of the present invention is as shown in FIG.

In FIG. 11, the Doppler filter calculation unit 7 is provided independently of the position information filter calculation unit (wake specification calculation unit 5), and based on the time series of the Doppler speed of the detection data correlated with the existing track, The smooth value and the predicted value of the Doppler velocity are calculated and input to the wake related data 4C.
In this case, the Doppler speed narrowing process (step S31) in FIG. 3 uses the predicted value of the Doppler filter calculated in the Doppler filter calculation process (step S125) described later, and among the Doppler speed candidates, the reliability Narrowing is performed by selecting the candidate closest to the distance change rate of the wake with the highest wake.

In the distance change rate coincidence determination process (step S32 in FIG. 3), the Doppler speed uniquely selected in the previous narrowing process (step S31) and the distance change rate obtained from the predicted value of the wake are calculated. The degree of coincidence is statistically calculated using the covariance of both errors.
In step S32, if it is determined that the Doppler speed of the detection data and the distance change rate of the wake do not match (that is, NO), the combination of both is rejected because there is no correlation between the detection data and the existing wake (step S32). S33) If it is determined that they match (that is, YES), a gate inside / outside determination process using the position information (step S34) and a correlation likelihood calculation process (step S35) are executed. Is adopted (step S36).

を用いて行われる。
このとき、ドップラフィルタの状態変数は、以下の式（３１）のように、ドップラ速度とドップラ速度の変化率とによる２つの変数で構成される。 In the Doppler filter calculation process (step S125) in FIG. 12, the update calculation of the Doppler filter is performed using the observed value of the Doppler velocity.

It is done using.
At this time, the state variable of the Doppler filter is composed of two variables based on the Doppler speed and the change rate of the Doppler speed, as in the following Expression (31).

をドップラフィルタの平滑値とし、

を平滑誤差共分散行列、

をカルマンゲイン、

を残差共分散行列とすると、ドップラフィルタの平滑値、平滑誤差共分散行列、カルマンゲインおよび残差共分散行列は、それぞれ、以下の式（３２）〜（３５）で表される。 Hereinafter, a method for calculating the smoothed value and the predicted value of the Doppler filter will be described.
First, for the smooth value of the Doppler filter,

Is the smooth value of the Doppler filter,

Smoothing error covariance matrix,

The Kalman gain,

Is the residual covariance matrix, the smoothed value of the Doppler filter, the smoothing error covariance matrix, the Kalman gain, and the residual covariance matrix are expressed by the following equations (32) to (35), respectively.

は観測行列、

は観測誤差共分散行列であり、また、

は予測値、予測誤差共分散行列である。 here,

Is the observation matrix,

Is the observation error covariance matrix, and

Is a prediction value and a prediction error covariance matrix.

  On the other hand, the predicted value of the Doppler filter is expressed by the following equations (36) and (37).

は推移行列、

は駆動雑音共分散行列である。 here,

Is the transition matrix,

Is the driving noise covariance matrix.

As described above, according to the fifth embodiment of the present invention, by comparing the “Doppler speed based on the Doppler frequency” and the “Doppler speed based on the Doppler shift difference”, a high-quality Doppler speed is calculated and used for tracking. can do.
Further, by constructing a filter based on the Doppler speed and narrowing down candidates compared with the predicted value of the Doppler filter, the correlation between the detection data and the existing track can be accurately determined.

Embodiment 6 FIG.
In the fifth embodiment (see FIG. 11), the predicted value is input from the track-related data 4C to the gate inside / outside determination unit 31 and the gate inside / outside determination processing is executed using the predicted value. Similarly to steps 3 and 4, the gate inside / outside determination processing may be executed using a smooth value.
The sixth embodiment of the present invention in which the gate inside / outside determination processing is executed using the smooth value will be described below.
FIG. 13 is a block diagram showing the sixth embodiment of the present invention, and FIG. 14 is a flowchart showing the individual correlation determination process according to the sixth embodiment of the present invention, which is the same as described above (see FIGS. 11 and 8). For the above, the same reference numerals as those described above are attached, and detailed description thereof is omitted.

13 is different from the above (FIG. 11) only in that it includes a gate inside / outside determination unit 31B using a smooth value and wake-related data 4D for inputting a smooth value to the gate inside / outside determination unit 31B.
14 differs from the above (FIG. 8) only in that a Doppler filter calculation process (step S141) is added instead of the wake smoothing value calculation process (step S81).
The tracking process procedure according to the third embodiment of the present invention is as shown in FIG. 2, and the Doppler speed information extraction process (step S22) is the same as described above.

In FIG. 13, a Doppler filter calculation unit 7 using a time series of Doppler velocities is provided independently of the position information filter calculation unit (wake specification calculation unit 5), as described above (element 11).
FIG. 14 shows detailed processing for determining individual correlations.
Hereinafter, the processing content of each step in FIG. 14 will be described.
First, in the Doppler filter calculation process (step S141), the smoothing value of the Doppler filter is updated with the Doppler speed candidate. At this time, since there are a plurality of Doppler speed candidates, a plurality of updated smooth values are also calculated.

Subsequently, in the Doppler speed narrowing process (step S31B), the distance change rate calculated using the predicted value of the track is compared with the smoothed value of the Doppler filter calculated by the Doppler filter calculation process (Step S141). To do.
At this time, narrowing down is performed by uniquely selecting a Doppler velocity candidate used for calculating the smooth value of the nearest Doppler filter.

Next, in the distance change rate coincidence determination process (step S32B), the degree of coincidence between the Doppler speed uniquely selected in step S31B and the distance change rate obtained from the predicted value of the wake is obtained, and the covariance of both errors is calculated. Use to calculate statistically.
In step S32B, if it is determined that the Doppler speed of the detection data matches the distance change rate of the existing track (that is, YES), then, using the position information as in the conventional device, (Step S34) and correlation likelihood calculation (step S35).

  On the other hand, if it is determined in step S32B that the Doppler speed of the detection data does not match the distance change rate of the existing track (that is, NO), it is assumed that there is no correlation between the detection data and the existing track, and this combination Is rejected (step S33).

As described above, according to the sixth embodiment of the present invention, by comparing the “Doppler speed based on the Doppler frequency” and the “Doppler speed based on the Doppler shift difference”, a high-quality Doppler speed is calculated and used for tracking. can do.
Further, by constructing a filter based on the Doppler speed and narrowing down candidates compared with the predicted value of the Doppler filter, it is possible to accurately determine the correlation between the detection data and the existing track.

It is a block block diagram which shows Embodiment 1 of this invention. It is a flowchart which shows the schematic procedure of the tracking process by Embodiment 1 of this invention. It is a flowchart which shows the individual correlation determination process by Embodiment 1 of this invention. It is a block block diagram which shows Embodiment 2 of this invention. It is a flowchart which shows the schematic procedure of the tracking process by Embodiment 2 of this invention. It is a flowchart which shows the individual correlation determination process by Embodiment 2 of this invention. It is a block block diagram which shows Embodiment 3 of this invention. It is a flowchart which shows the individual correlation determination process by Embodiment 3 of this invention. It is a block block diagram which shows Embodiment 4 of this invention. It is a flowchart which shows the schematic procedure of the tracking process by Embodiment 4 of this invention. It is a block block diagram which shows Embodiment 5 of this invention. It is a flowchart which shows the schematic procedure of the tracking process by Embodiment 5 of this invention. It is a block block diagram which shows Embodiment 6 of this invention. It is a flowchart which shows the schematic procedure of the tracking process by Embodiment 6 of this invention. It is explanatory drawing which shows the specific correlation process relevant to Embodiment 1-6 of this invention.

Explanation of symbols

  1 sensor, 2 Doppler velocity information extraction unit, 3 individual correlation determination unit, 31, 31B gate inside / outside determination unit, 32 likelihood calculation unit, 4, 4B, 4C, 4D track related data, 5 track specification calculation unit, 6, 6A Tracking track determination unit, 7 Doppler filter calculation unit.

Claims (6)

A target tracking device that obtains position information and Doppler velocity from detection data obtained by observation with a pulse compression radar, and tracks a target using an observation value of the position information and an observation value of the Doppler velocity,
A Doppler speed information extraction unit that narrows down first Doppler speed candidates from the detection data;
A gate internal / external determination unit that determines a correlation of the detection data with respect to an existing track from the detection data and the first Doppler speed;
A tracking track calculation unit that performs calculation for determining the tracking track of the target using the existing track and the detection data;
The Doppler velocity information extraction unit obtains the first Doppler velocity candidate from the frequency deviation between the transmission frequency and the reception frequency of the pulse compression radar, and the distance obtained by alternately executing up-chirp and down-chirp. Calculating a second Doppler velocity from the distance deviation of the observed values, comparing the first and second Doppler velocities, and narrowing down the first Doppler velocity candidates;
The gate inside / outside determination unit compares the predicted speed of the existing track with a plurality of candidates for the first Doppler speed obtained from the Doppler speed information extraction unit, and uniquely determines the first Doppler speed candidate. It is determined whether or not the detection data and the existing track can be correlated with each other based on the speed deviation between the predicted speed and the uniquely selected first Doppler speed candidate. A target tracking device, wherein a possible area of the detection data is calculated from a predicted position, and it is determined whether or not the detection data and the existing track can be correlated. A target tracking device that obtains position information and Doppler velocity from detection data obtained by observation with a pulse compression radar, and tracks a target using an observation value of the position information and an observation value of the Doppler velocity,
A Doppler speed information extraction unit that narrows down first Doppler speed candidates from the detection data;
A gate internal / external determination unit that determines a correlation of the detection data with respect to an existing track from the detection data and the first Doppler speed;
A tracking track calculation unit that performs calculation for determining the tracking track of the target using the existing track and the detection data;
The Doppler velocity information extraction unit obtains the first Doppler velocity candidate from the frequency deviation between the transmission frequency and the reception frequency of the pulse compression radar, and the distance obtained by alternately executing up-chirp and down-chirp. Calculating a second Doppler velocity from the distance deviation of the observed values, comparing the first and second Doppler velocities, and narrowing down the first Doppler velocity candidates;
The gate inside / outside determination unit obtains a predicted track speed with the highest reliability among existing tracks that can be correlated with the detection data, and a plurality of candidates for the first Doppler speed obtained from the Doppler speed information extraction unit. In comparison, the first Doppler speed candidates are uniquely narrowed down, and the detection data is compared with the predicted speed of the most reliable wake and the uniquely narrowed first Doppler speed. It is determined whether or not the existing track can be correlated, and the detection data existence area is calculated from the predicted position of the existing track, and whether or not the detection data and the existing track can be correlated is further determined. Judgment,
The tracking track calculation unit includes a likelihood calculation unit and a tracking track determination unit,
The likelihood calculating unit calculates a degree of correlation between the existing track and the detection data based on a degree of coincidence between the predicted speed of the most reliable track and the first Doppler speed,
The tracking track determination unit composes the target hypothesis by combining possible correlations with the existing track, and holds the target hypothesis based on the degree of correlation between the existing track and the detection data. A target tracking device that determines a tracking track of the target. A target tracking device that obtains position information and Doppler velocity from detection data obtained by observation with a pulse compression radar, and tracks a target using an observation value of the position information and an observation value of the Doppler velocity,
A Doppler speed information extraction unit that narrows down first Doppler speed candidates from the detection data;
A gate internal / external determination unit that determines a correlation of the detection data with respect to an existing track from the detection data and the first Doppler speed;
A tracking track calculation unit that performs calculation for determining the tracking track of the target using the existing track and the detection data;
The Doppler velocity information extraction unit obtains the first Doppler velocity candidate from the frequency deviation between the transmission frequency and the reception frequency of the pulse compression radar, and the distance obtained by alternately executing up-chirp and down-chirp. Calculating a second Doppler velocity from the distance deviation of the observed values, comparing the first and second Doppler velocities, and narrowing down the first Doppler velocity candidates;
The gate inside / outside determination unit includes a Doppler velocity smooth value included in the smooth value information calculated by updating the existing track with the detection data, and a first Doppler velocity information obtained from the Doppler velocity information extraction unit. Candidates are compared with each other, the speed deviation between the two closest to the Doppler speed smooth value among the first Doppler speed candidates is obtained, and the detection data and the existing track can be correlated from the speed deviation And further calculating whether or not the detection data and the existing track can be correlated by calculating a possible area of the detection data from the predicted position of the existing track. Tracking device. A target tracking device that obtains position information and Doppler velocity from detection data obtained by observation with a pulse compression radar, and tracks a target using an observation value of the position information and an observation value of the Doppler velocity,
A Doppler speed information extraction unit that narrows down first Doppler speed candidates from the detection data;
A gate internal / external determination unit that determines a correlation of the detection data with respect to an existing track from the detection data and the first Doppler speed;
A tracking track calculation unit that performs calculation for determining the tracking track of the target using the existing track and the detection data;
The Doppler velocity information extraction unit obtains the first Doppler velocity candidate from the frequency deviation between the transmission frequency and the reception frequency of the pulse compression radar, and the distance obtained by alternately executing up-chirp and down-chirp. A second Doppler velocity is calculated from the distance deviation of the observed values, and the first and second Doppler velocities are compared to narrow down the first Doppler velocity candidates;
The gate inside / outside determination unit determines the possibility of correlation based on a combination of the existing track and the detection data, and is calculated by updating a track having the highest reliability among the existing tracks that can be correlated with the detection data. By comparing the Doppler velocity smooth value included in the smoothed value information with a plurality of candidates for the first Doppler velocity obtained from the Doppler velocity information extraction unit, the first Doppler velocity candidates are uniquely narrowed down. Determining whether or not the detection data and the existing track can be correlated by comparing the Doppler velocity smooth value with the uniquely narrowed first Doppler velocity, and predicting the position of the existing track Calculating the possible area of the detection data from, further determining whether the detection data and the existing track can be correlated,
The tracking wake calculation unit includes a tracking wake determination unit,
The tracking track determination unit composes a plurality of hypotheses of the target by combining possible correlations with the existing track, and determines the plurality of hypotheses based on the degree of correlation between the existing track and the detection data. A target tracking device that determines a tracking track of the target while holding the target. A target tracking device that obtains position information and Doppler velocity from detection data obtained by observation with a pulse compression radar, and tracks a target using an observation value of the position information and an observation value of the Doppler velocity,
A Doppler speed information extraction unit that narrows down first Doppler speed candidates from the detection data;
A gate internal / external determination unit that determines a correlation of the detection data with respect to an existing track from the detection data and the first Doppler speed;
A Doppler filter calculation unit for calculating a smoothed value and a predicted value of the Doppler velocity;
A tracking track calculation unit that performs calculation for determining the tracking track of the target using the existing track and the detection data;
The Doppler velocity information extraction unit obtains the first Doppler velocity candidate from the frequency deviation between the transmission frequency and the reception frequency of the pulse compression radar, and the distance obtained by alternately executing up-chirp and down-chirp. Calculating a second Doppler velocity from the distance deviation of the observed values, comparing the first and second Doppler velocities, and narrowing down the first Doppler velocity candidates;
The gate inside / outside determination unit compares the predicted value of the Doppler velocity obtained from the Doppler filter with a plurality of candidates for the first Doppler velocity obtained from the Doppler velocity information extraction unit, and The Doppler speed candidates are uniquely narrowed down, it is determined whether or not the detection data and the existing track can be correlated, and the detection data existence area is calculated from the predicted position of the existing track, and the detection Further determine whether the data and the existing track can be correlated,
The Doppler filter calculation unit calculates a smooth value and a predicted value of the Doppler velocity based on a time series of Doppler velocity of detection data correlated with the existing track,
The tracking wake calculation unit includes a tracking wake determination unit,
The tracking track determination unit composes a plurality of hypotheses of the target by combining possible correlations with the existing track, and determines the plurality of hypotheses based on the degree of correlation between the existing track and the detection data. A target tracking device that determines a tracking track of the target while holding the target. A target tracking device that obtains position information and Doppler velocity from detection data obtained by observation with a pulse compression radar, and tracks a target using an observation value of the position information and an observation value of the Doppler velocity,
A Doppler speed information extraction unit that narrows down first Doppler speed candidates from the detection data;
A gate internal / external determination unit that determines a correlation of the detection data with respect to an existing track from the detection data and the first Doppler speed;
A Doppler filter calculation unit for calculating a smoothed value and a predicted value of the Doppler velocity;
A tracking track calculation unit that performs calculation for determining the tracking track of the target using the existing track and the detection data;
The Doppler velocity information extraction unit obtains the first Doppler velocity candidate from the frequency deviation between the transmission frequency and the reception frequency of the pulse compression radar, and the distance obtained by alternately executing up-chirp and down-chirp. A second Doppler velocity is calculated from the distance deviation of the observed values, and the first and second Doppler velocities are compared to narrow down the first Doppler velocity candidates;
The gate inside / outside determination unit compares the smoothed value of the Doppler velocity obtained from the Doppler filter with the plurality of candidates for the first Doppler velocity obtained from the Doppler velocity information extraction unit, The Doppler speed candidates are uniquely narrowed down, it is determined whether or not the detection data and the existing track can be correlated, and the detection data existence area is calculated from the predicted position of the existing track, and the detection Further determine whether the data and the existing track can be correlated,
The Doppler filter calculation unit calculates a smooth value and a predicted value of the Doppler velocity based on a time series of Doppler velocity of detection data correlated with the existing track,
The tracking wake calculation unit includes a tracking wake determination unit,
The tracking track determination unit composes a plurality of hypotheses of the target by combining possible correlations with the existing track, and determines the plurality of hypotheses based on the degree of correlation between the existing track and the detection data. A target tracking device that determines a tracking track of the target while holding the target.

Images