KR20040055430A - Position Controller And Controlling Method for Time Delay Compensation of the Image Tracker in Electro-Optical Tracking System - Google Patents

Abstract

PURPOSE: A time delay compensating controller of an image tracker in an electro-optical tracking system(EOTS) and a control method thereof are provided to construct a smith predictor for operating independently relating to an input command and a disturbance input within a loop, thereby controlling effectively the input command follow-up and disturbance compensation function. CONSTITUTION: An image sensor unit(230) is comprised of a subtraction unit(211) for subtracting a chase output signal(y) from the input command. A quantizer(212) quantizes the output of the subtraction unit(211). An image processing unit(220) is comprised of an adder unit(221) for adding the output of the image sensor unit(210) to a quantization error signal(n), a sampling period unit(222) and a laplace converting unit(223) of an image process time delay. A control input(u) and a disturbance(d) of a position controller(240) are added in the adder unit(250) and inputted it to a chase loop. A stabilization drive unit(230) outputs an output signal(y) of a chaser.

Description

Position Controller And Controlling Method for Time Delay Compensation of the Image Tracker in Electro-Optical Tracking System

The present invention relates to tracking control technology of an Electro-Optical Tracking System (EOTS), and more particularly, to minimize the effect of the time delay of the image sensor used in the electro-optical tracker on the tracking loop performance. The present invention relates to a time delay compensation controller and a control method.

In general, an electro-optical tracker (EOTS) provides a stable line of sight to an observer while automatically tracking targets moving in the presence of disturbances such as vibrations in aircraft, such as ships, tanks and helicopters. ) And target information such as target position, velocity, and angular velocity to the fire control system.

The electro-optical tracker (EOTS) is usually composed of an image sensor unit, an image tracking unit, and a stabilization driving unit. The stabilization driving unit is composed of a mechanism supporting the image sensor and a servo control unit moving the mechanism. The servo control unit is composed of a motor, a servo amplifier, a gyroscope for stabilization, and hardware and software for combining them.

As described above, the stabilization driving unit of the EOTS is composed of all mechanical devices and electronic control devices for tracking based on the position value of the target calculated from the image tracking unit. Therefore, it is accompanied by the difficulty of simultaneously controlling mechanical devices including non-linear effects as well as electronic parts that are relatively ideally modeled and calculated, and time due to disturbance and image processing due to mechanical vibration generated when the tracker is moved. Delays will greatly affect the performance of the entire electro-optical tracker.

Among the various problems of the electro-optical tracker, the tracking image, which is directly perceived by the observer, has a time delay due to the operation principle, and it is known that the effect of the time delay characteristic on the entire system has a significant effect on the tracking loop performance.

In order to solve this problem, a performance improvement method using a Smith predictor has been proposed in the conventional position controller design. The Smith predictor was originally known as a general control loop design method for systems with time delays, but the crucial disadvantage of Smith predictors is that step-by-step disturbances can occur when a time-delayed plant includes a pure integral mode. When applied, the output of the system will no longer follow the target value.

Therefore, various modified Smith predictors based on the Smith predictor have been actively studied in order to solve these problems. The modified Smith predictor is called Dead Time Compensator (DTC) in other terms and has been studied by Watanabe, Matausek, Normey-rico and others.

On the other hand, the precision of stabilization gimbal in the electro-optical tracker is very high. In the existing system, the attenuation constant of gimbal has a non-negligible effect on the whole system. High precision machining is now possible. In the past, low-precision machining was used to control the tracking loop by constructing a position controller using PI, PID, or lead-lag controller, which is a classic control method.

However, the problem using the Smith predictor, which assumes that the speed loop does not have the ideal operating characteristics as the problem that has not been considered as described above is possible to achieve the ideal speed loop control by the realization of high precision machining, Problems have arisen that are not applicable.

In order to make up for the shortcomings of the Smith predictor, Matausek et al. Developed the DTC, but the controller design had to be changed according to the amount of time delay of the plant, and it was not able to generate the optimal control signal for the disturbance input. If this change, there is a problem that can not guarantee the stability of the system.

That is, in the case of a robot running by using an image tracking system, when a delay occurs in a plant, a closed loop system is generally designed using a Smith predictor. However, this method has a disadvantage in that an error occurs in the output of the disturbance input when the plant has a pure integration mode, so there is a problem that cannot be effectively applied in a system such as an electro-optical tracker in which the disturbance input exists.

The present invention is to solve the above problems, and has a Smith predictor structure that operates independently for the input command and the disturbance input in the loop, which is the most important performance index in the electro-optical tracker (EOTS). An object of the present invention is to provide a time delay compensation controller and a control method of an image tracking unit in an electro-optical tracker that enables effective control of input command tracking and disturbance compensation functions.

As such, the controller structure used in the present invention is capable of setting independent control constants when setting input commands and disturbance compensation control constants when designing a control loop, and has the advantage of always ensuring stability of the system.

In the electro-optical tracker of the present invention for achieving the above object, a time delay compensation controller of an image tracking unit includes an image sensor unit, an image processing unit, a position controller, and a stabilization driving unit. and a disturbance observation model for optimally arranging a Smith predictor in response to (r) and a disturbance input signal (d) to form a loop, and a position controller auxiliary control constant portion.

In addition, the method of controlling the time delay compensation of the image tracking unit in the electro-optical tracker of the present invention, in the target tracking method using the electro-optical tracker, in response to the input command signal (r) and the disturbance input signal (d) respectively Smith (Smith) When the unit step input is applied by adding and controlling the disturbance observation model and the position controller auxiliary control constant part to the predictor, the input command r converges to the target value and the disturbance input d is removed. It features.

1 is a block diagram of a general configuration of an electro-optical tracker to which the present invention can be applied;

2 is a block diagram of an entire system of an electro-optical tracker (EOTS) to which the present invention is applied;

3 is a block diagram of a position control loop of an electro-optical tracker (EOTS) in accordance with the present invention;

4A is a graph illustrating response characteristics of an input command and disturbance among response characteristics of a tracking loop having a conventional PI controller;

Figure 4b is a graph showing the tracking error of the response characteristics of the tracking loop having a conventional PI controller,

5A is a graph illustrating response characteristics of an input command and disturbance among response characteristics of a tracking loop having a controller according to the present invention;

5b is a graph illustrating tracking error in response characteristics of a tracking loop having a controller according to the present invention;

Explanation of symbols on main parts of the drawing

210: image sensor unit 220: image processing unit

230: stabilization drive unit 240, 320: position controller

340: integral plant with time delay

401: input waveform 402: output waveform

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of an overall configuration of an electro-optical tracker (EOTS) to which the present invention can be applied.

Referring to the drawing, an electro-optical tracker (EOTS) obtains target information (Position / Velocity Information) from an image sensor unit (10) and image data for obtaining an image signal, and stabilizes the image while tracking the eye. An image tracker 20 and a stabilization driver 30 are included. In this case, the stabilization driver 30 is again sensed by the image sensor unit 10 through optical coupling.

2 is an overall system block diagram of an electro-optical tracker (EOTS) to which the present invention is applied, and shows an image sensor 210, an image processor 220, a position controller 240, and a stabilization driver 230.

The image sensor 210 includes a subtractor 211 which subtracts the tracking output signal y from the input command r according to the target movement, a quantizer 212 that quantizes the output of the subtracter 211, and a sensor. An image processing unit 220 includes an adder 221 for adding a quantization error signal n to an output of the image sensor unit 210, and a sampling period unit 222. And the Laplace transform unit 223 of the image processing time delay.

And the control input (u) and the disturbance (d) of the position controller 240 is added in the adder 250 is input to the tracking loop, the stabilization driver 230 is the equivalent model unit 231, speed-position of the stabilization loop The conversion dynamics section 232 and the position sensor gain section 233 output the tracker output signal y.

Referring to FIG. 2, the image sensor unit 210 receives an image from a TV camera or a thermal image sensor, converts the image into digital image data, and provides the image to the image processor 220. ) Processes the image data to recognize and track the target and transmits the information to the stabilization driver 230. The stabilization driver 230 drives the image sensor unit 210 by using the position or velocity information coming from the image processor 220 to continuously track the target.

Stabilization driver 230 refers to a device for moving the image sensor unit 210 in the target direction by receiving a position error of the target and the origin provided by the image tracking unit, the target tracking driver for precise tracking is a stabilization loop therein It also includes a 'stabilization driver'.

That is, the stabilization driver 230 tracks the target based on the signal applied from the image tracking unit, and compensates the disturbance (d) generated during the movement of the moving object equipped with the electro-optical tracker (EOTS) to stabilize the target surroundings. It is responsible for providing a more stable observation environment to the observer by providing an eye.

Here, the electro-optical tracker (EOTS) uses a gyroscope to detect angular velocity or angle information for detecting disturbance around gimbal, and stabilizes the gaze by having the detected angular velocity or angle information in the feedback loop. To keep track of your targets.

3 is a block diagram of a position control loop of an electro-optical tracker (EOTS) in accordance with the present invention.

The target tracking sequence of the electro-optical tracker (EOTS) first achieves high image quality by operating the stabilization system. In addition, the operator operates in the manual mode of capturing and recognizing the target while searching and monitoring the target. When the target is selected, information on the target is obtained and the target is fixed to enter the tracking mode.

Referring to FIG. 3, the input command signal r according to the target movement and the feedback output of the position control loop are input to the position controller 320 through the subtractor 301, and the position controller 320 is a position controller control signal. The control signal compensated for the disturbance compensation signal (^ d) to (Ur), and in addition to the disturbance input (d) in the adder 330, the control signal is removed from the influence of the disturbance (d) has a time delay Is applied to the integral plant 340.

Here, the position controller 320 according to the present invention is a modified Smith predictor in which a position controller auxiliary control constant (Ko) 309 and a disturbance observation model 310 are added to the structure of the Smith predictor.

On the other hand, the relationship between FIG. 2 and FIG. 3 can be understood by comparing the positions of various input signals such as an input command r, a disturbance input d, and a quantization error input n. The relationship can be understood by the description of Equation 1 below.

Equation 1 is equivalent to expressing the sum of the sensor sampling time delay 213 and the image processing time delay 223, the speed control loop 231, and the position conversion function 232, which are time delays of the electro-optical tracker. G_p ^ * (s) may be expressed as Equation 2 below.

Variable time delay T in the equation (3) indicates a total time delay amount, which is calculated simply as the sum of the image sensor sampling delay time (T _s) and the video signal time delay (T _d) of the processor 220. In addition, to indicate the estimated value of τ is a delay time T from the plant equation (4), the delay time in the amount of electro-optical tracker T because it is the nature of the system constant τ uses the T value as it is. In addition, since the gain estimate K _p of the plant can know the actual gain K of the plant, it can be set to the same value as in Equation 5 above.

As shown in FIG. 3, when the disturbance observation model 310 and the position controller auxiliary control constant 309 are added to design the position controller 320 of the electro-optical tracker according to the present invention, the input command signal r The optimal compensation signal ^ d is generated for the disturbance input d as well as the response characteristic of the C.sub.2, thereby eliminating the time delay effect of the EOTS.

To this end, the present invention constitutes a block 310 that optimally arranges a Smith predictor in response to the input command signal r and the disturbance input signal d to form a loop, thereby ensuring independent operation. Each response characteristic can be adjusted using different constants.

That is, referring to the relationship with the position controller 240 of FIG. 2, the block diagram of the position control loop according to the present invention is shown in FIG. 3 to improve the performance of the position control loop.

If the plant and the model satisfy the equations (4) and (5), the transmission characteristics of the electro-optical tracker using the control method according to the present invention are derived by the following equations (6) and (7).

The poles of the denominator of Equation 7 all exist at the origin and the left half plane. Therefore, using the control method according to the invention it can be seen that the entire system is always stable regardless of the time delay value of the plant.

When the unit step input is applied to the input command r and the disturbance input d, the final value theorem is applied to Equations 6 and 7, respectively, as shown in Equation 8 below. It can be seen that the input command r converges to the target value. In addition, as shown in Equation 9, it can be seen that the disturbance input d is removed.

As shown in Equation 6, the response time constant T _r of the output for the input command of the proposed DTC is given by 1 / (K _p K _r ). Therefore, the settling time of the output for the unit step input command can be expressed as T + 5 / (K _r K _p ).

Deriving the disturbance compensation output (^ d) for the disturbance input d to obtain the response speed of the disturbance compensation output (^ d) to the disturbance input (d) is given by the following equation (10).

As shown in Equation 10, it is confirmed that the proposed method can accurately output the disturbance compensation signal after the time T + 5 / (K _O K _p ). One thing to note here is that, as can be seen in the denominators of Equations 6 and 10, the method of the present invention can set the control variables independently regardless of the magnitude of the time delay of the tracker.

In conclusion, using the method according to the present invention, the convergence rate of the output for the input command of the plant with time delay can be determined by the control variable K _r , and the generation rate of the disturbance compensation signal for the disturbance input is K _o . The convergence speed can be adjusted independently of the K _r value.

In addition, as can be seen in the denominator of Equation 7, since no variables cause stability problems in the setting of K _r and K _o variables, the selection of control variable values can be freely set according to the required performance. This means that when using the control method according to the present invention, the tracking loop can be designed regardless of the time delay performance of the image tracking unit of the electro-optical tracker.

In order to confirm the validity of the control method according to the present invention, a simulation was performed using Matlab, the control loop and control system simulation software of FIG. In the simulation, the speed gain constant K _p is used as the position gain constant of the resolver used as a position sensor in the actual system, 25.4, which is compared with the result of using the method according to the present invention and the PI control method. Among the control variables used in the PI control method, the gain of the speed gain P was 100 and the integral gain I value was 30. In the method according to the present invention, only the constant gain was used as the K _r value was 100.

In addition, in the waveforms of FIGS. 4A, B and 5A, B showing the simulation, the solid line represents the input command, and the dotted line represents the output of the plant. For the input waveform, the signals represented by the acceleration movement section, the constant speed calculation, and the deceleration movement section are used to represent the movement of the moving object, and the maximum amplitude is ± 1 and the waveform has a period of 4 seconds. In addition, in order to determine the disturbance compensation characteristics, the disturbance (d) value was input as -0.5 value in FIG. 3 after 0.1 second after the start of the simulation.

As shown in FIG. 4A, when the PI controller is used, the output waveform is delayed due to the time delay characteristic of the plant before 0.1 second, and when the disturbance is input after 0.1 second, the system cannot accurately compensate the disturbance that is input. It can be seen that if the output waveform has a large error compared to the input waveform, it will be traced. The tracking error is shown in Figure 4b.

However, when using the control method according to the present invention as shown in Figure 5a it can be seen that even if the output waveform disturbance input to the input waveform has a superior control characteristics compared to the method using the PI controller. This is because the method according to the present invention has a structure capable of compensating for the input disturbance as described in Equation 10. As shown in the tracking error characteristic shown in FIG. 5B, the method has better tracking error characteristics than the conventional method. You can check it.

These results can be used as a result of solving the time delay problem of the image sensor in the design and implementation of the next-generation electro-optical tracker.For the tracking system based on the video system such as unmanned surveillance system, unmanned guard system, unmanned road driving equipment, etc. It may also be applicable.

What has been described above is only one embodiment that describes a time delay compensation controller and a control method of an image tracking unit in an electro-optical tracker according to the present invention, and the present invention is not limited to the above-described embodiment, and the following claims Without departing from the gist of the present invention claimed in the scope, anyone of ordinary skill in the art to which the present invention extends to the technical idea of the present invention to the extent that various modifications can be made.

Claims (4)

In the electro-optical tracking device having an image sensor unit, an image processing unit, a position controller, and a stabilization driving unit, The position controller, A disturbance observation model for optimally arranging a Smith predictor in response to the input command signal r and the disturbance input signal d, respectively, to form a loop, and a position controller auxiliary control constant portion. Time Delay Compensation Controller of Image Tracer in Electro-optical Tracker. The method of claim 1, wherein the transmission characteristics of the electro-optical tracker and And a unit step input is applied to the input command r and the disturbance input d, the input command r converges to a target value and the disturbance input d is removed. Time Delay Compensation Controller of Image Tracker in Optical Tracker. 3. The disturbance compensation output (^ d) for the disturbance input d is calculated by the equation of claim 2, in order to obtain a response speed of the disturbance compensation output (d) for the disturbance input (d). Time delay compensation controller of the image tracker in the electro-optical tracker, characterized in that defined as. In the target tracking method using an electro-optical tracker, When the unit step input is applied by adding and controlling the disturbance observation model and the position controller auxiliary control constant part to the Smith predictor in response to the input command signal r and the disturbance input signal d, respectively, And the input command (r) converges to a target value, and the disturbance input (d) is removed.