JP2003248057A - Radar-tracking apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a radar-tracking apparatus that reduces loss and miss- tracking of an object to be tracked, even if the fleet of the object to be tracked such as a ship or the like moves under a clutter environment. <P>SOLUTION: In the radar-tracking apparatus that has correlation processing means, tracking filter means, position information processing means, and prediction processing means, and tracks a plurality of moving targets, a group-tracking means for grouping a plurality of targets for tracking is provided. Then, group information is compared with the tracking information of the tracking filter means, appropriate information is selected for each object to be tracked, and appropriate track data are created by the position information processing means. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a radar tracking device for tracking a plurality of moving targets such as ships.

[0002]

2. Description of the Related Art A moving target such as a ship is tracked by using a radar tracking device. FIG. 8 is a diagram showing a configuration of a radar tracking device for a marine vessel that has been generally used conventionally. In FIG. 8, the transmitter 2-1
The radar antenna 1 irradiates the ship target with a transmission signal from the ship as a microwave. The reflected wave from the target is received by the radar antenna 1, converted into a received signal by the receiver 2-2, and sent to the signal processor 3. The signal processor 3 detects target information such as the position and size of the target from the received signal and sends it to the data processor 4. This target information is an observed value obtained from the radar reception signal, and includes position data including distance / azimuth, speed data, target size data, and the like. The data processor 4 extracts the target information of the tracking target,
Track data in which position information for each tracking target is arranged in time order is created and sent to the radar display 5. The radar display 5 displays the track data on the radar screen.

FIG. 9 is a diagram showing the configuration of a data processor 4 which performs tracking processing. Based on target information (distance, azimuth angle, speed, etc.) of the tracking target, the true value (distance, azimuth) of the tracking target is shown. Horn,
(Speed etc.) is estimated and arranged in time series to create track data. The true value is a value obtained by removing the error from the target information, but since it is actually unobservable, a smoothed value is usually applied.

In FIG. 9, the target information from the signal processor 3 is first input to the correlation processing means 41. Here, the correlation processing is performed between the track data prediction information (position and speed) created in the previous scan and the target information of the current scan input from the signal processor 3. Scan refers to one rotation of the radar antenna when the radar tracking device is operating in TWS (Track While Scan), and therefore target information of the same target is obtained for each scan.

The correlated target information is sent to the tracking filter means 42 as track data correlation information. The tracking filter means 42 smoothes the track data correlation information to obtain the track data smoothing information. The track data smoothing information is input to the position information processing means 43 and the prediction processing means 44. Position information processing means 43
Then, the track data position information is calculated. The track data position information is sent to the radar display 5 as track data and stored as track data. Further, this track data position information is sent to the prediction processing means 44. The prediction processing means 44 calculates the track data prediction information for the next scan, and the correlation processing means 4
1, supplied to the tracking filter means 42.

As an automatic tracking algorithm for a plurality of targets in a clutter environment, NNT (Neares) is used for tracking type correlation processing.
t Neighbor Tracking) method and PDA (Probabilistic D)
ataAssociation) method is known.

The operation of the data processor 4 when the NNT system is used for the correlation processing will be described with reference to the explanatory view of FIG. In the correlation process of FIG. 9A, a correlation gate is set around the prediction information (predicted position) predicted by the prediction processing unit 44 during the previous scan (scan number: k-1). The correlation gate is set in a range in which the presence of the target is predicted around the predicted position of the tracking target (estimated value of the target position for the next scan (scan number: k)). In this example,
Of the target information X1 to X6, target information X1, X2, X3
Exists in the correlation gate.

Then, as shown in FIG. 2B, for all the target information X1, X2, X3 entered in the correlation gate,
The distances from the predicted position are compared, and the closest target information X2 is set as the tracking target track data correlation information. Next, the tracking filter means 42 uses the track data prediction information of the current scan k and the like to smooth the track data correlation information. By this smoothing processing, track data smoothing information to be tracked is obtained.

In the position information processing means 43, as shown in FIG. 3C, the current scan (scan) is performed by the track data position information of the previous scan (scan number: k-1) and the input track data smoothing information. The track data position information of the scan number: k) is calculated. The calculated track data position information is stored as track data and sent to the radar display 5 as track data. At the same time, the track data position information is calculated by the prediction processing means 44.
Is input to. The prediction processing circuit calculates the predicted position of the next scan (scan number: k + 1) in consideration of the data input time interval (scan rate) as shown in FIG.

The operation when the PDA method is used for the correlation processing will be described below.
Set a correlation gate centered on the predicted position. Then, a weighting coefficient is given to all the target information entered in the correlation gate from the correlation probability, and the weighted average target information is calculated. This weighted average target information is used as track data correlation information. Subsequent processing is performed similarly to the NNT method.

[0011]

However, in the conventional tracking method, when a fleet or the like composed of a plurality of tracking targets is tracked, the lost tracking target and the mistracking of the tracking target increase.

That is, when the NNT method is used for the correlation processing, the size of the correlation gate is adjusted to maintain the tracking accuracy. However, if the correlation gate size is made large, the existence probability of the tracking target is high and the lost of the tracking target is reduced, but the mistrack of the tracking target is increased. On the contrary, if the correlation gate size is made small, the mistrack of the tracking target is reduced, but the existence probability of the tracking target is low and the lost of the tracking target is increased. Therefore, it is difficult to maintain tracking accuracy by adjusting the correlation gate size. Further, in the PDA method, which has improved the correlation processing method under the clutter environment, the correlation gate size can be set large to reduce the lost of the tracking target, but the fleet or the like composed of a plurality of closely tracking targets is tracked. In this case, it is difficult to separate and track the respective tracking targets, and thus it is difficult to reduce the mistracks of the tracking targets.

In view of the above problems, the present invention provides a radar tracking device that reduces the lost tracking target and the mistracked tracking target even when the tracking target such as a ship moves collectively in a clutter environment. The purpose is to

[0014]

According to a first aspect of the present invention, there is provided a radar tracking device, comprising: correlation processing means for extracting track data correlation information for each tracking target from target information obtained by processing a received signal and track data prediction information; and the track data correlation. Tracking filter means for extracting smoothed track data smoothing information in consideration of an error of the track data correlation information from the information and the track data prediction information, and the track data including the position information, to which the track data smoothing information is input. Position information processing means for calculating information, outputting at least the position information as display data, and outputting the track data information for prediction processing, and the track based on the track data information from the position information processing means. Data prediction information is created, and the correlation processing means and the tracking filter are created. A radar tracking device for tracking a plurality of moving targets, which comprises a prediction processing means for supplying the target information, and further receives the target information, and groups the plurality of targets based on a moving state including an interval distance and a moving direction. Group tracking means that outputs the smoothed group information that is smoothed for each group, and is provided between the tracking filter means and the position information processing means. The track data smoothing information and the smoothing information are provided. Generalization group information is input, the input track data smoothing information is compared with the smoothing group information, and appropriate information is selected for each tracking target, and the position information processing is performed as track data optimization information. And target information comparing means for inputting to the means.

A radar tracking device according to a second aspect of the present invention is the radar tracking device according to the first aspect, wherein the group tracking means sets a plurality of targets on the basis of the distances and the advancing directions and advancing speeds of individual targets which can be known from radar Doppler signals and the like. Group defining means for dividing into groups by group, group centroid detecting means for detecting the group centroid value by referring to all target information belonging to the group for each group, and included in the group centroid value and group prediction information Group correlation processing means for extracting the correlation group center of gravity value for each group from the center of gravity predicted position, and the smoothed group information smoothed in consideration of the error of the correlation group center of gravity value from the correlation group center of gravity value and the group prediction information. And the group tracking filter means for supplying the target information comparing means to the target information comparing means and the smoothing means. Loop position information is input, group position information is calculated, group position information processing means for outputting for prediction processing, and group prediction information based on the group position information and the smoothed group information from this group position information processing means. And group prediction processing means for supplying the group correlation processing means and the group tracking filter means.

A radar tracking device according to a third aspect of the present invention is the radar tracking device according to the first aspect.
2. In the radar tracking device according to any one of 2 above, the target information comparing unit selects a group to which a specific target belongs from the smoothing group information, track data smoothing information of the specific target, and the specific target. Smoothing information comparing means for comparing the smoothing group information to which the target belongs and outputting a selection signal of which one to select, track data smoothing information of the specific target, and the smoothing to which the specific target belongs Target information switching means for switching any one of the group information according to the selection signal and outputting it as track data optimization information.

The radar tracking device according to claim 4 is the radar tracking device according to claim 3, wherein the track data smoothing information and the smoothing group information used in the smoothing information comparing means are velocity data. Is characterized by.

The radar tracking device according to a fifth aspect is the radar tracking device according to the third aspect, wherein the track data smoothing information and the smoothing group information used in the smoothing information comparing means are acceleration data. Is characterized by.

A radar tracking device according to a sixth aspect is the radar tracking device according to the third aspect, wherein the track data smoothing information and the smoothing group information used in the smoothing information comparing means are velocity data and acceleration data. Is characterized in that.

The radar tracking device according to claim 7 is the radar tracking device according to any one of claims 1 to 3.
In the radar tracking device according to any one of 6 above, in the target information comparing means, an absolute value of a difference between the track data smoothing information and the smoothing group information is a predetermined comparison coefficient (hereinafter, a predetermined ratio). When it is larger than the product of the smoothed group information, the smoothed group information is output as track data optimum information, and conversely, when it is smaller, the track data smoothed information is output as track data optimum information. And

The radar tracking device according to claim 8 is the radar tracking device according to claim 1.
In the radar tracking device according to any one of 7), the target information comparing means is set to output the smoothed group information as track data optimum information in an initial stage of tracking.

According to the radar tracking device of the present invention, when the tracking target such as a plurality of ships moves as a group (fleet), it is possible to prevent mistracking of other targets by the same scan. When the possibility of mistracking is high, the target information comparing means selects the smoothed group information, and the prediction processing means in the next stage predicts the target information for the next scan from the smoothed group information. As a result, the correlation processing of the next scan can be performed with the same tracking target existence probability as other tracking targets in the group, and it is not necessary to increase the correlation gate size. If the correlation gate size is constant, the occurrence of mistrack will be suppressed, and the tracking process will be stable.

If the target information does not exist in the correlation gate, the track data in the current scan will be lost. In the present invention, since the group information is interpolated by the prediction processing means, the tracking processing is performed with the missing data as it is. A stable tracking process becomes possible rather than executing.

Further, in the initial tracking process after the activation of the radar tracking device, the track data smoothing information to be tracked is often unstable, but when it is determined that the track data smoothing information is not stable (for example, tracking For a certain time after the start or when the smoothing target information changes drastically),
The smoothing group information is used in the prediction processing means. In the initial tracking process, the track data smoothing information gradually approaches the smoothing group information, and when it becomes stable, the initial tracking process shifts to the operation tracking process, the smoothing target information becomes stable, and the tracking accuracy remains constant. As the existing radar tracking device, it is possible to start creating track data.

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to FIG.
~ It demonstrates in detail with reference to FIG.

FIG. 1 is a diagram showing the configuration of a data processor 4A of the radar tracking device of the present invention. This data processor 4
A receives the target information from the signal processor 3, performs data processing, and outputs track data to the display 5.
In its internal configuration, the correlation processing means 41, the tracking filter means 42, the position information processing means 43, and the prediction processing means 44.
Is similar to the conventional radar tracking device of FIG.

In the present invention, the group tracking means 50 is newly added.
And the target information comparison means 60 are added. The group tracking means 50 creates smoothed group information based on the target information and sends it to the target information comparison means 60. The target information comparing means 60 compares the track data smoothing information with the smoothing group information, selects one of them and outputs it as the track data optimizing information to the position information processing means 43.

FIG. 2 is a diagram showing the structure of the group tracking means 50. The group definition means 51, the group center of gravity detection means 52, the group correlation processing means 53, the group tracking filter means 54, and the group position information processing. Means 5
5 and group prediction processing means 56.

The group defining means 51 divides a plurality of targets into groups based on the moving state such as the distance, the traveling direction, and the traveling speed. This group definition process is divided into three stages as shown in FIG. At the first stage, the target information such as position, size and Doppler velocity is received as shown in FIG. At the second stage, as shown in FIG. 9B, the group definition distance, the group definition speed, and the group definition traveling direction are examined from the target information. At the third stage, as shown in FIG. 7C, the target information that meets the group definition conditions is defined as the same group. There are also individual goals that are not defined in groups.

The group centroid detecting means 52 detects the group centroid value and the group size for each group by referring to all the target information belonging to the group.

The group correlation processing means 53 extracts the correlation group center of gravity value for each group from the group center of gravity value from the group center of gravity detecting means 52 and the center of gravity predicted position included in the group prediction information from the group prediction processing means 56. Specifically, when the NNT method is used for the correlation processing, the correlation gate is set centering on the prediction information (center of gravity predicted position) predicted by the group prediction processing means 56 in the previous scan (scan number: k-1). . This correlation gate is set in a range in which the existence of the barycentric value of the group is predicted around the barycenter predicted position of the tracking group. Then, the distance from the predicted position of the center of gravity is compared, and the center of gravity value of the closest group is set as the center of gravity value of the correlation group of the tracking target.

In the group tracking filter means 54, smoothed group information obtained by smoothing the correlation group centroid value and the group prediction information (centroid predicted position) from the group prediction processing means 56 in consideration of the error of the correlation group centroid value. Take out.

The group position information processing means 55 receives the smoothed group information, calculates the smoothed position information including the group position information, and outputs it for the prediction process. In addition,
The smoothed group information supplied to the target information comparison means 60 may be supplied via the group position information processing means 55, or may be supplied directly from the group tracking filter means 54.

FIG. 4 is a diagram showing the configuration of the target information comparing means 60, which receives the track data smoothing information from the tracking filter means 42 and the smoothing group information from the group tracking means 50, and receives either of them. Select and send as track data optimization information. The track data smoothing information is for each individual tracking target and includes the track number, the target number used in the correlation processing, the individual speed, and the individual acceleration. The smoothed group information is for each group and includes a group number, a target number to which it belongs, a group speed, and a group acceleration. The target information comparing means 60 includes a group selecting means 61 for selecting a group to which the target belongs, a speed / acceleration comparing means 62 for comparing speed and acceleration as smoothing information, and a target information switching means 6
3 and 3.

The group selecting means 61 to which the target belongs selects a group to which the target having the track data smoothing information belongs from a plurality of groups, and smoothing group information (that is, group speed Vg, group) of the group. The acceleration Ag) is supplied to the speed / acceleration comparison means 62 and the target information switching means 63. On the other hand, the target track data smoothing information (that is, the track data smoothing speed V
t, track data smoothing acceleration At) are also supplied to the speed / acceleration comparing means 62 and the target information switching means 63.

The velocity / acceleration comparing means 62 takes the difference between the track data smoothing information and the smoothing group information regarding the same target, and the absolute value of the difference exceeds a predetermined ratio of the smoothing group information or is less than that. It is determined whether or not it is supplied to the target information switching means 63 as a selection signal. The target information switching means 63 outputs the smoothed group information as the track data optimum information when the absolute value of the difference exceeds the predetermined ratio of the smoothed group information,
On the contrary, if it does not exceed, the track data smoothing information is output as the track data optimum information.

A data processor 4A having the above-mentioned configuration
The operation of is as follows.

Based on the target information from the signal processor 3,
The correlation processing means 41 and the tracking filter means 42 can obtain track data smoothing information for each individual tracking target.
Here, the track data smoothing speed Vt is assumed as the track data smoothing information.

On the other hand, at the same time, the group tracking means 50 obtains smoothed group information for each group based on the target information from the signal processor 3. Here, the group speed is assumed. From among these multiple group speeds,
The group to which the target belongs is selected and the group speed Vg of that group is obtained. With respect to the group speed Vg, the speeds of the individual tracking targets in the group are averaged, and the speed change for each scan becomes gentle. Similarly, for other observation values, the observation error of each tracking target is averaged and reduced.

In the speed / acceleration comparing means 62, the absolute value of the difference between the target smoothing speed Vt and the group speed Vg is compared with a predetermined ratio kv times the group speed Vg.
Expressed as an expression, it is determined whether | Vt−Vg |> kv · Vg or | Vt−Vg | <kv · Vg.

In the former case, since the absolute value of the difference exceeds the predetermined ratio kv of the group speed Vg, the group speed Vg is output as the track data optimization information.
In the latter case, since the absolute value of the difference does not exceed the predetermined ratio kv of the group speed Vg, the smoothing speed Vt is output as the track data optimization information.

An example of the operation of this speed comparison process is shown in FIG. In this FIG. 5, three goals are defined as the same group. The mark ○ indicates the track data for each scan. The thick solid line indicates the target smoothing speed V
t, and the thin solid line indicates the group velocity Vg.

As can be seen from FIG. 5, when the difference between the smoothing speed Vt and the group speed Vg is small, the followability is taken into consideration so as to correspond to the gradual change in the track data smoothing information. , Track data smoothing information (smoothing speed Vt, smoothing acceleration At) is selected. On the other hand, when the difference between the smoothing velocity Vt and the group velocity Vg is large, the smoothing group information (group velocity Vg, group acceleration Ag) is regarded as a rapid change caused by the influence of clutter and other targets. select.

In this way, the track data smoothing information (smoothing speed V
t, smoothed acceleration At) or smoothed group information (group velocity Vg, group acceleration Ag), the track data information calculated by the position information processing means 43 is
As well as being supplied to the display 5 as track data,
The track data prediction information for the next scan is created by being supplied to the prediction processing means 44.

In the speed / acceleration comparing means 62, smoothed acceleration A is used as track data smoothing information.
t, the group acceleration Ag is used as the smoothing group information, and | At−Ag |> ka · Ag or | At−
By determining whether Ag | <ka · Ag, it is possible to operate similarly to the case of speed. Note that ka
Is a predetermined ratio.

Further, the smoothing velocity Vt and the smoothing acceleration At may be used as the track data smoothing information, and the group velocity Vg and the group acceleration Ag may be used as the smoothing group information. In the selection process in this case, for example, the smoothing speed Vt and the group speed Vg are first calculated.
, | Vt-Vg |> kv · Vg, or | V
It is determined whether t-Vg | <kv · Vg. | Vt-
When Vg |> kv · Vg, the smoothed group information (group velocity Vg, group acceleration Ag) is used as the track data optimization information.

On the other hand, if │Vt-Vg│ <kvVg, then using the smoothing acceleration At and the group acceleration Ag, whether │At-Ag│> kaAg or │At-
It is determined whether Ag | <ka · Ag. | At-Ag
In the case of |> ka · Ag, the smoothed group information (group velocity Vg, group acceleration Ag) is used as the track data optimization information, and | At−Ag | <ka · Ag
In the case of, track data smoothing information (smoothing speed V
t, smoothing acceleration At) is used as the track data optimization information.

FIG. 6 is a diagram showing an example of track data of the radar tracking device according to the present invention, and FIG. 6 (a) shows the relationship between the radar antenna 1 and the three vessels A, B and C to be tracked. There is. The three ships A, B, and C are moving at equal intervals W at an average navigation speed. As shown in FIG.
It can be seen that the track data A, B, C are accurately created as shown by the solid line. On the other hand, FIG. 7 is a diagram showing track data by the conventional radar tracking method under the same conditions for comparison. As shown in FIG. 7, in the conventional radar tracking device, the tracking target A,
In the track data of B and C, it can be seen that a tracking target miss track and a tracking target lost have occurred from an early stage of tracking.

Further, in the initial tracking process after the activation of the radar tracking device, the track data smoothing information to be tracked is generally unstable in many cases. In order to cope with such a problem in the initial tracking process, when it is determined that the track data smoothing information is not stable, the target information comparing means 60 selects the smoothing group information,
The smoothing group information is used to make a prediction in the prediction processing means 44. In this case, in the initial tracking process, when the track data smoothing information gradually approaches the smoothing group information and becomes stable, the initial tracking process shifts to the operation tracking process, and the smoothing target information becomes stable,
As a radar tracking device that maintains a constant tracking accuracy, it is possible to start creating track data.

Various methods can be adopted as a method for selecting the smoothed group information in the initial tracking process. As a concrete means, for example, a speed difference (Vt
-Vg) or the difference between accelerations (At-Ag) in determining the magnitude, proportionality coefficients kv and ka are made small immediately after the start of tracking to facilitate selection of smoothing group information, and then set to a predetermined value. Method, a method of forcibly selecting smoothed group information for a certain period of time after the start of tracking, or monitoring fluctuation (fluttering) of track data smoothed information,
While the fluctuation is large, a method of selecting smoothed group information or the like can be adopted.

[0051]

According to the present invention, by using the track data optimization information for a fleet composed of a plurality of targets, the prediction error can be reduced and the correlation gate size can be reduced by the correlation processing. It is possible to reduce the occurrence of mistracks to be tracked. Also, if the target information does not exist in the correlation gate and the track data is missing,
Since the interpolation is performed using the group information, the tracking process is stable, and the lost tracking target can be reduced.

Further, the present invention is effective in a process such as the initial tracking process where the track data smoothing information is not stable. Further, a re-tracking process of re-creating track data can be executed even for a target that is out of the tracking target and is not conventionally created with the track data while belonging to the group. That is, it has the effect of keeping the tracking performance of the radar tracking device constant.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a data processor of a radar tracking device of the present invention.

FIG. 2 is a diagram showing a configuration of a group tracking unit in the data processor of the present invention.

FIG. 3 is a diagram illustrating group definition processing.

FIG. 4 is a diagram showing a configuration of target information comparing means in the data processor of the present invention.

FIG. 5 is a diagram showing an operation example of speed comparison processing.

FIG. 6 is a diagram showing track data of the radar tracking device according to the present invention.

FIG. 7 is a diagram showing track data according to a conventional radar tracking method under the same conditions.

FIG. 8 is a diagram showing a configuration of a radar tracking device for a marine vessel, which is generally used in the past.

FIG. 9 is a diagram showing the configuration of a conventional data processor that performs tracking processing.

FIG. 10 is a diagram illustrating an operation of a conventional data processor.

[Explanation of symbols]

1 radar antenna 2-1 transmitter 2-2 Receiver 3 signal processor 4, 4A data processor 41 Correlation processing means 42 Tracking filter means 43 Position Information Processing Means 44 Prediction processing means 5 Radar display 50 group tracking means 51 Group definition means 52 Group center of gravity detection means 53 Group correlation processing means 54 group tracking filter means 55 Group Position Information Processing Means 56 group prediction processing means 60 Target information comparison means 61 Group selection processing means to which the target belongs 62 Speed / acceleration comparison means 63 Target information switching means

Continued front page    F term (reference) 5J070 AC02 AC06 AC07 AE02 AH04                       AH19 AK15 AK16 AK22 BB04                       BB05 BB06 BB07 BB16

Claims (8)

[Claims] 1. Correlation processing means for extracting track data correlation information for each tracking target from target information obtained by processing a received signal and track data prediction information, and track data correlation information from the track data correlation information and the track data prediction information. Tracking filter means for extracting the smoothed track data smoothing information in consideration of the error of the track data, the track data smoothing information is input, track data information including position information is calculated, and at least the position information is used as display data. Position information processing means for outputting the track data information for prediction processing, and for producing the track data prediction information based on the track data information from the position information processing means. And a prediction processing unit that supplies the tracking filter unit. In a radar tracking device that tracks a number of moving targets, further receives the target information, divides a plurality of targets into groups based on a moving state including a moving distance and a moving direction, and smooths each group. Group tracking means for outputting smoothing group information, provided between the tracking filter means and the position information processing means, the track data smoothing information and the smoothing group information are inputted, and the inputted Comparing the track data smoothing information and the smoothing group information,
A radar tracking device, comprising: target information comparing means for selecting appropriate information for each tracking target and inputting it to the position information processing means as track data optimization information. 2. The group tracking means includes a group defining means for dividing a plurality of targets into groups based on a moving state including a distance and a moving direction, and for each group, all target information belonging to the group. Group centroid detection means for detecting the group centroid value with reference, group correlation processing means for extracting the correlation group centroid value for each group from the group centroid value and the centroid predicted position contained in the group prediction information, the correlation group centroid value And from the group prediction information,
Taking out the smoothed group information smoothed in consideration of the error of the correlation group center of gravity value, the group tracking filter means for supplying to the target information comparison means, and the smoothed group information is input, and group position information is input. A group position information processing unit that calculates and outputs for prediction processing, and creates the group prediction information based on the group position information and the smoothed group information from this group position information processing unit, and the group correlation processing unit and The radar tracking apparatus according to claim 1, further comprising a group prediction processing unit that supplies the group tracking filter unit. 3. The target information comparing unit selects a group to which a specific target belongs from the smoothing group information, track data smoothing information of the specific target, and the smoothing group to which the specific target belongs. Smoothing information comparing means for comparing the information and outputting a selection signal indicating which one to select, and one of the specific target track data smoothing information and the smoothing group information to which the specific target belongs. 3. The radar tracking device according to claim 1, further comprising target information switching means for switching according to the selection signal and outputting as track data optimization information. 4. The radar tracking device according to claim 3, wherein the track data smoothing information and the smoothing group information used by the smoothing information comparing means are velocity data. 5. The radar tracking device according to claim 3, wherein the track data smoothing information and the smoothing group information used by the smoothing information comparing means are acceleration data. 6. The radar tracking device according to claim 3, wherein the track data smoothing information and the smoothing group information used by the smoothing information comparing means are velocity data and acceleration data. 7. The target information comparing means sets an absolute value of a difference between the track data smoothing information and the smoothing group information between a predetermined comparison coefficient (hereinafter, a predetermined ratio) and the smoothing group information. If the product is larger than the product, the smoothed group information is output as the track data optimum information, and conversely, if it is smaller, the track data smoothed information is output as the track data optimum information.
The radar tracking device according to claim 1. 8. The target information comparing means is set so as to output the smoothed group information as track data optimum information at an initial stage of tracking. The radar tracking device according to Crab.