KR20150053605A - Ownship motion control method for enhancement of target motion analysis performance - Google Patents

Abstract

The present invention relates to an own ship motion control method capable of improving the accuracy of target tracking by maximizing observability to a target, comprising the steps of: detecting own ship data and target data; selecting possible directions in which a target is positioned within the sensor coverage of the own ship, avoiding geographic features among several progressing directions which can veer from the current location of the own ship; and maneuvering the own ship in the direction maximizing the observability to the target among the selected directions.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method and apparatus for controlling a target maneuver,

The present invention relates to a maneuvering maneuvering maneuver, and more particularly, to a maneuvering maneuvering control method capable of improving the accuracy of target tracking by maximizing the maneuverability of a target.

Target motion analysis, which estimates the position of a target based on azimuthal information or Doppler-shifted frequency measurement of a target, has been studied for a long time in various fields, such as sensor-based public or underwater surveillance come.

Therefore, it is necessary to change ownership in order to improve the performance of target maneuver analysis that tracks the target with azimuth information. If the ambiguity changes, the rate of change of the azimuth angle increases and the visibility increases, resulting in accurate tracking of the target.

However, although the number of operations can empirically be used to judge the direction in which the observability increases, it is possible that such judgment is inaccurate. There are many researches on the self - recommendation maneuver which increases the maneuverability because it can reduce the burden of the operation number by automating the self - maneuvering process.

 Generally, since the sensor standard deviation of each sensor mounted on the robot is different, it is necessary to take into account the standard deviation of each orientation when the robot manages the robot. That is, it is advantageous for the position of the target to be in the bearing with a small sensor standard deviation during self-starting. There is an area where the sensor can reach according to the position where the sensor of the human body is installed. This area is called the sensor coverage area.

We did not consider the sensor coverage area in the existing self - test. However, if the target deviates from this sensor coverage area when the ambiguity changes, the problem of missing the target occurs. In addition, the direction in which you hit the feature should be avoided.

Accordingly, it is an object of the present invention to provide a self-recommendation start-up control method capable of improving the target maneuvering performance by maximizing the visibility of the target.

Another object of the present invention is to provide a self-recommendation start-up control method capable of maximizing the visibility of the target while considering the sensor coverage area, the standard deviation of the sensor orientation, and the feature.

According to another aspect of the present invention, there is provided a method of controlling a self-recommendation start-up, the method comprising: detecting target information and user identification information; Selecting directions in which the estimated target is located within the sensor coverage of the feature to avoid among the various directions of travel from the current position of the subject; And activating the target in a direction that maximally increases the spectral response to the target among the selected directions.

Wherein the step of selecting the direction in which the estimated target is located comprises the steps of: obtaining sensor coverage reachable by the sensor at a predicted position of each object moving for a predetermined period of time from a position of the current object to a direction in which the object is displaceable; Obtaining estimated positions of the targets at the predicted positions of the respective objects; And selecting a predicted position in which the sensor coverage includes an estimated target position among the predicted positions of the target and avoids the feature.

The step of activating the autonomy may include calculating an azimuth angle with respect to the target estimated position at an angle formed by two tangents tangent to the distance and azimuth reliability ellipse of the target and a predicted position of each box at a predicted position of the subject moving for a predetermined time in each progressive direction, Obtaining a star sensor standard deviation; And determining a predicted position of the object to be maximized and the sensor standard deviation to be minimized in a direction that maximally increases the objectivity; And activating the engagement in the determined direction.

The method of claim 1, wherein, after the target has been moved for a predetermined period of time, the target is moved in the direction that maximizes the latency to the target using the distance between the future target and the future target and the bearing reliability ellipse .

According to another aspect of the present invention, there is provided a self-recommendation start-up control method for obtaining a sensor coverage reachable by a sensor at each predicted position moved for a predetermined time from a position of a current owner step; Obtaining an estimated position of the target at each of the predicted positions; Selecting the prediction locations in which the obtained sensor coverage includes an estimated target location and avoids the feature; And activating the robot in a direction that maximally increases the sensibility of the target in consideration of the standard deviation of the sensor and the feature at the selected predicted position.

Wherein the step of activating the autonomy comprises: drawing two tangents in contact with the distance and the bearing confidence ellipse of the target at the predicted position of each autonomy; Obtaining a bearing standard deviation of each bearing by obtaining a bearing with respect to the target bearing position at a predicted position of each bearing; And determining a position where the angle formed by the two tangential lines is maximized and the sensor standard deviation is minimized at the predicted position of the character, and activating the ruler in the direction of the determined position.

The method of claim 1, wherein, after the target has been moved for a predetermined period of time, the target is moved in the direction that maximizes the latency to the target using the distance between the future target and the future target and the bearing reliability ellipse .

The self-recommendation start-up control method according to the present invention manages the target in the direction of maximizing the probabilistic response to the target in consideration of the sensor coverage, the standard deviation of the sensor according to the orientation, and the feature, There is an effect that can be done.

In addition, the self-recommendation start-up control method according to the present invention can also be used as an algorithm for automatically controlling a torpedo in order to secure visibility of a target.

1 is a block diagram showing a self-test start-up configuration;
Figure 2 shows a typical sensor coverage area of the present submarine.
Fig. 3 is a diagram showing a geometrical relationship between an estimated position of a target and a character; Fig.
4 is a diagram showing a geometrical relationship between an estimated position of a target and a character;
FIG. 5 is a diagram showing a geometric relationship between a predicted estimated position of a future target and a character; FIG.

FIG. 1 is a block diagram illustrating an operation of the self-recommendation start control proposed in the present invention.

As shown in FIG. 1, the arithmetic operation unit calculates the target maneuver target (TMA) filter based on the azimuth angle of the target input at every time step and the charter information (position, speed) After initializing the particle, it outputs the estimation information (position, velocity, distance uncertainty, bearing uncertainty) of the target.

Distance and Bearing Reliability The ellipse forming part forms a distance and bearing reliability ellipse centered on the target based on the target estimation information and the sensor information detection part detects the sensor coverage area and the feature information of the subject Output.

Based on the target estimation information, distance and azimuth reliability ellipses are formed as follows. The relative estimated distance (R), bearing uncertainty (STD_B), and distance uncertainty (STD_R) for the TMA result target can be obtained. Distance and bearing reliability ellipses indicate areas where the target may exist above a certain probability. The length of the major axis in this ellipse is STD_R, and the length of the minor axis is R * tan (STD_B). As shown in Fig. 4 and Fig. 5, the long axis of the distance and azimuth reliability ellipse is the direction from the estimated target position toward the current position, and the short axis is the direction perpendicular to the long axis. The center of the ellipse is the estimated position of the target. Once the lengths of the long and short axes are obtained, the vertices of each axis are connected to form an ellipse.

The control unit (or control unit) selects the directions in which the estimated target is located in the sensor coverage of the object to avoid the object among the various directions of travel that can be changed from the current position of the object, and then, The maneuvering is started in the direction of increasing.

Although the operation of each of the above-described detecting units has been described separately, the present invention is not limited to this, and the user information detecting unit and the sensor information detecting unit can be integrated into one. Also, the operation of the distance and azimuth reliability ellipsoid forming unit and the sensor information detecting unit may be performed in the self-recommendation start-up control unit.

)을 설정한다. 각 진행 방향으로 일정 시간(t_p)동안 움직인 후 예상되는 n개 자함의 위치를 찾는다. 상기 t_p는 표적까지 추정되는 거리에 비례하고 자함의 속도에 반비례하도록 설정한다. 모든 i≤n에 대하여 v_i방향으로 t_p시간 동안 이동 후 자함의 예상위치를 p_i라고 한다. The control unit (or control unit) can control the direction of movement

). Find the position of the expected n characters after moving for a certain time (t _p ) in each direction of travel. T _p is proportional to the estimated distance to the target and is set to be inversely proportional to the speed of the object. For every _i ≤ n, the expected position of the movement after the movement in the direction v _i is t _p , is called p _i .

There is an area where the sensor can reach according to the position where the sensor of the human body is installed. This area is called the sensor coverage area.

Figure 2 is a typical sensor coverage area of a submarine viewed from above.

If the target is out of the sensor's coverage area of the subject, tracking of the target is missed. Thus, out of each position p _i ( _i ≤ n), where the robot moves for t _p , only the positions where the sensor coverage includes the estimated target position are searched.

Fig. 3 shows the geometrical relationship between the estimated position of the target and the target.

As shown in FIG. 3, the position of the current frame is indicated by a green dot and the estimated position of the target is indicated by a red dot. Since the direction that can be changed from the position of the current box is eight directions, eight arrows are displayed based on the box.

The predicted position p _i ( _i ≤ n) of the subpixel after t _p time can be represented by the end point of each arrow, and when reaching any predicted position, the sensor coverage area appears as two sectors. In particular, FIG. 3 depicts locations where the sensor coverage does not include the estimated target among the predicted positions p _i ( _i ? N) of the target.

Therefore, if the target moves in the direction depicted in FIG. 3, the estimated target (green point) deviates from the sensor coverage of the fan-shaped sensor. Therefore, the direction depicted in FIG.

Also, the maneuver must be maneuvered in a direction to avoid collision with the feature. Therefore, we need to identify locations where a straight line connecting each location p _i ( _i ≤ n) avoids features at the current location of the resource.

로 나타내면

중에서 최적의 자함 기동 방향은 다음 방법으로 찾을 수 있다. Accordingly, the invention is primarily jaham yi _p t time sensor coverage at each location p _i (i≤n) moving the target position includes an estimation filter out during the position avoiding the feature (select). These locations

As shown in FIG.

The optimal maneuver direction can be found in the following way.

optimal Self  How to find the start direction

에서 표적의 거리 및 방위 신뢰도 타원 (error ellipse)에 두 접선을 긋는다. 두 접선 사이의 이루는 각이 클수록 해당 위치로 자함이 이동했을 때 가관측성이 증가함을 의미한다. When the sensor coverage contains the estimated target location and the locations avoiding the feature are obscured,

, Draw two tangents to the distance and bearing confidence ellipse of the target. The larger the angle between the two tangents, the greater the visibility of the robot when the robot moves to that position.

에서 표적의 거리 및 방위 신뢰도 타원에 접하는 두 접선이 이루는 각을

로 표현한다. 각 예측 위치

에서 표적 추정 위치에 대한 방위각을 구한 후 이 방위에 대한 센서 표준편차를

로 나타낸다. 자함은

를 최대화시키며 동시에

를 최소화하는 방향으로 기동해야 한다. 이를 위하여 다음 수학식 1과 같은 비용함수(cost function :

)을 정의한다. Each predicted position

The distance of the target and the azimuth of the target.

. Each predicted position

And the sensor standard deviation for this orientation

Respectively. The

At the same time

Should be minimized. To this end, a cost function (cost function:

).

[Equation 1]

는 0보다 크거나 같은 변경 인자(tunning factor)이다.

를 0으로 설정하면 방위별 센서 표준 편차를 고려하지 않은 경우이며,

가 커질수록 방위별 센서 표준편차를 고려한 함수가 된다. 각각의 위치

에 대하여 해당하는

를 찾고

가 최대가 되는 방향으로 자함을 기동한다. 물론, 수학식 1이 아니어도

를 최대화시키며 동시에

를 최소화하는 비용함수가 존재한다면 사용 가능하다(예:

). In the above formula

Is a tuning factor greater than or equal to zero.

Is set to 0, the standard deviation of the sensor is not taken into consideration,

Is a function that takes into account the standard deviation of the sensor by bearing. Each location

Against

Looking for

In the direction of the maximum. Of course,

At the same time

Lt; RTI ID = 0.0 > (e. G., ≪ / RTI &

).

Fig. 4 shows the geometric relationship between the estimated position of the target and the target, and shows the optimum firing direction for the estimated target.

는 각 화살표의 끝점으로 표시되었다. 자함의 예측 위치 중에서

가 가장 큰 위치는 도 4에 묘사되었다. 추정 표적의 위치(붉은 점)가 도 4에 묘사된 자함의 예측 위치에서 센서 커버리지(두 개의 파선 부채꼴)안에 존재하므로, 도 4에 나타난 방향으로 자함 권고 기동이 이루어진다.The position of the subject box is indicated by a green dot and the target position is indicated by a red dot. The distance and bearing confidence ellipse of the target is depicted as an ellipse centered on the target. Since the directions that can be changed from the position of the present box are eight directions, eight arrows are displayed on the basis of the box. Predicted location of character after t _p time

Is indicated by the end of each arrow. Among the predicted positions

The largest position is depicted in FIG. Since the position (red dot) of the estimated target exists in the sensor coverage (two dashed sectors) at the predicted position of the character depicted in FIG. 4, the self-recommendation maneuver is performed in the direction shown in FIG.

The present invention considers the sensor coverage, the standard deviation of the sensor per orientation, and the topographic features, and suggests a self - recommendation maneuver to maximize the visibility of the target. Since the self-recommendation method proposed in the present invention is simple and requires only a small amount of additional computation, it is suitable for real-time self-recommendation maneuvering, and thus can be used as an automatic control system of a torpedo in order to secure the visibility of the target.

In addition, the present invention can be extended to a method of calculating the self-recommendation maneuver in consideration of the future position / speed of the target. That is, on the basis of the target start-up results obtained so far, the target is with respect to t _p time is after the future target position and a future target distance and orientation confidence ellipse (predicted error ellipse) of the predicted movement for the cost function of Equation 1 (cost function is moved in the direction of maximizing the movement.

FIG. 5 is a geometric relationship diagram of the predicted estimated position of the future target and the target, and shows the optimal maneuver direction for the predicted target.

)가 가장 큰 위치가 도 5에 묘사되어 있다. Referring to FIG. 5, the position of the subject box is indicated by a green dot and the estimated position of the current target is indicated by a red dot. Since the target is moved for time t _p future target position has been displayed in red dots (50), which is a broken line. And the predicted distance of the target and the predicted error ellipse are described as ellipses centering on the future target position. Therefore, the cost function (1) of Equation (1) with respect to the future target position, the distance between the predicted future target and the predicted error ellipse,

) Are shown in Fig.

As described above, according to the present invention, it is possible to track a target more accurately with a simple and small amount of computation by activating the robot in the direction of maximizing the probabilistic response to the target in consideration of the sensor coverage, the standard deviation of the sensor according to the orientation, There are advantages.

The above-described self-recommendation start-up method according to the present invention described above is not limited to the configuration and method of the embodiments described above, but the embodiments may be embodied in other specific forms without changing their technical ideas or essential features It will be understood that the invention may be practiced. Therefore, it should be understood that the above-described embodiments are to be considered in all respects as illustrative and not restrictive.

50: Future target location

Claims (7)

Detecting target information and user information;
Selecting directions in which the estimated target is located within the sensor coverage of the feature to avoid among the various directions of travel from the current position of the subject; And
And activating the target in a direction that maximally increases the spectral response to the target among the selected directions. 2. The method of claim 1, wherein selecting a direction in which the estimated target is located
Obtaining a sensor coverage reachable by the sensor at a predicted position of each of the compartments moving for a predetermined time in a measurable direction from the position of theidentity;
Obtaining estimated positions of the targets at the predicted positions of the respective objects; And
And selecting a predicted position where the sensor coverage includes an estimated target position among the predicted positions of the target and avoids the feature. 2. The method of claim 1, further comprising: activating the charter;
Obtaining a sensor standard deviation of the azimuth angle with respect to the target estimated position at an angle formed by two tangents tangent to the distance and azimuth reliability ellipse of the target at a predicted position of the object moving in each advanceable direction for a predetermined time; And
Determining a predicted position of a person whose standard deviation is minimized and a calculated angle to be the maximum, in a direction that maximally increases the observability; And
And activating the self-driving in the determined direction. 2. The method of claim 1, further comprising: moving the target in a direction that maximizes the latency to the target using the distance between the future target and the future target and the bearing reliability ellipse after the target moves for a predetermined time The method comprising the steps of: Obtaining a sensor coverage reachable by the sensor at each predicted position that has been moved for a predetermined time in a direction in which the sensor can be turned from the position of the current compartment;
Obtaining an estimated position of the target at each of the predicted positions;
Selecting the prediction locations in which the obtained sensor coverage includes an estimated target location and avoids the feature; And
And activating the robot in a direction that maximally increases the spectral response to the target in consideration of the standard deviation of the sensor and the feature at the selected predicted position. 6. The method of claim 5, wherein activating the self-
Drawing two tangents in contact with the distance and azimuth reliability ellipse of the target at the predicted position of each jaw;
Obtaining a bearing standard deviation of each bearing by obtaining a bearing with respect to the target bearing position at a predicted position of each bearing; And
And determining a position at which the angle formed by the two tangential lines is maximized and the standard deviation of the sensor is minimized at the predicted position of the magnetic field, and activating the magnetometer in the direction of the determined position. 6. The method of claim 5, further comprising: after the target has been moving for a predetermined time, activating the target in a direction that maximally increases the target distance using the distance between the future target position and the future target and the bearing reliability ellipse And a control unit for controlling the start-up control method.

Images