• Proceedings of the KIEE Conference
• /
• 2006.10c
• /
• pp.266-268
• /
• 2006

In this paper, we design a Linear Quadratic Gaussian(LQG) controller for active suspensions. We can improve the inherent suspension problem, trade-off between the ride quality and the suspension travel by selecting appropriate weights in the LQ-objective function. Using an optimization-algorithm, Evolution Strategy(ES), we find the proper control gains for selected frequencies, which have major effects on the vibrations of the vehicle's state variables. The frequencies and proper control gains are used for the neural network data. During a vehicle running, the trained on-line neural network is activated and provides the proper gains for non-trained frequencies.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1530-1533
• /
• 2005

In this paper, we design a Linear Quadratic Gaussian controller for the active suspension. We can improve the inherent suspension problem, trade-off between the ride quality and the suspension travel by selecting appropriate weights in the LQ-objective function. Because any definite rules for selecting weights do not exist, we use an optimization-algorithm, Evolution Strategy (ES) to find the proper control gains for selected frequencies, which have major effects on the vibrations of the vehicle's state variables. The frequencies and proper control gains are used for the neural network data. During a vehicle running, the trained on-line neural network is activated and provides the proper gains for non-trained frequencies. For the full-state feedback control, Kalman filter observes the full states and Fourier transform is used to detect the frequency of the road.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.3
• /
• pp.263-272
• /
• 2000

In this paper we give some geometric condition for a stochastic nonlinear system and we propose a control method for a stochastic nonlinear system using neural networks. Since a competitive learning neural networks has been developed based on the stochastcic approximation method it is regarded as a stochastic recursive filter algorithm. In addition we provide a filtering and control condition for a stochastic nonlinear system called the perfect filtering condition in a viewpoint of stochastic geometry. The stochastic nonlinear system satisfying the perfect filtering condition is decoupled with a deterministic part and purely semi martingale part. Hence the above system can be controlled by conventional control laws and various intelligent control laws. Computer simulation shows that the stochastic nonlinear system satisfying the perfect filtering condition is controllable and the proposed neural controller is more efficient than the conventional LQG controller and the canonical LQ-Neural controller.

• Proceedings of the KSME Conference
• /
• 2001.06b
• /
• pp.93-98
• /
• 2001

This paper studies on the design of an ILQ(Inverse Linear Quadratic optimal control) looper control system for hot strip mills. The looper which is placed between each stand plays an important role in controlling strip width by regulating strip tension variation generated from the velocity difference of main work rolls. The mathematical model for looper is firstly obtained by Taylor's linearization of nonlinear differential equations, where it is given as a linear and time invariant state-space equation. Secondly, a looper servo controller is designed by ILQ control algorithm, which is an inverse problem of LQ(Linear Quadratic optimal control) control. By tunning control gain arbitration parameters and time constants, it is shown that the ILQ looper servo controller has the performance that makes well to follow desired trajectories of both strip tension and looper angle.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.346-352
• /
• 1998

In this paper, we give some geometric condition for a stochastic nonlinear system and we propose a control method for a stochastic nonlinear system using neural networks. Since a competitive learning neural networks has been developed based on the stochastic approximation method, it is regarded as a stochastic recursive filter algorithm. In addition, we provide a filtering and control condition for a stochastic nonlinear system, called perfect filtering condition, in a viewpoint of stochastic geometry. The stochastic nonlinear system satisfying the perfect filtering condition is decoupled with a deterministic part and purely semi martingale part. Hence, the above system can be controlled by conventional control laws and various intelligent control laws. Computer simulation shows that the stochastic nonlinear system satisfying the perfect filtering condition is controllable. and the proposed neural controller is more efficient than the conventional LQG controller and the canoni al LQ-Neural controller.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.8
• /
• pp.1680-1689
• /
• 2002

This paper studies on the design of a looper control system for hot strip mill finisher using ILQ(Inverse Linear Quadratic optimal control) control method. The loopers are placed between each rolling stands and looper control plays an important role in regulating strip tension. The strip tension is controlled by raising and lowering the looper and by changing the speed of main work rolls. Firstly, it is shown from a nonlinear dynamic simulation that the strip tension is more influenced by difference of rolling speed than that of the looper angle. Secondly, a servo controller of the looper is designed using ILQ control method of which the characteristics and algorithms are simply introduced. Finally, the performances of the ILQ servo controller are compared with those of the LQI servo controller from computer simulation. In result, it is shown that the proposed ILQ servo controller has the better performances and robustness far parameter perturbations and disturbances than those of LQI controller.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.8
• /
• pp.145-155
• /
• 1998

An $H{\infty}$ controller with a roll eccentricity filter is proposed to alleviate the effect of entry thickness variation and roll eccentricity occured in rolling stands themselves of tandem cold mills. A robust controller to the disturbances is designed by H$_{\infty}$ control techniques, which can reflect the input direction of disturbances and the knowledge of disturbance spectrum in the frequency domain. First, fundamental problems in tandem cold mills such as process transport delay inherent in the exit thickness measurement and the feedforward loading of roll eccentricity signals on the exit thickness be overcome by the roll eccentricity filtering and the compensation for the error of gaugemeter thickness estimator. And non-satndard $H{\infty}$ control problem caused by the selection of weighting function having poles on the $J{\omega}$-axis is discussed. The resultant controller composed by an $H{\infty}$ controller and an estimator for the roll eccentricity is evaluated through computer simulations. The effectiveness of the proposed control method is compared to that of the conventional LQ controller method and a feedforward controller for the roll eccentricity, which has been already studied.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.2
• /
• pp.243-251
• /
• 2003

In this paper, a new model for mobile soccer robot, a type of linear system, is presented. A controller, consisting of two loops the one of which is the inner state feedback loop designed for stability and plant be well conditioned and the outer loop is a well-known PI controller designed for tracking the reference input, is suggested. Because the plant, the soccer robot, is parameter dependent, it requires the controller to be insensitive to the parameter variation. To achieve this objective, the pole-sensitivity as a pole-variation with respect to the parameter variation is defined and design algorithms for state-feedback controllers are suggested, consisting of two matrices one of which is for general pole-placement and other for parameter insensitive. This paper shows that the PI controller is equivalent to the state feedback and the cost function for reference tracking is equivalent to the LQ cost. By using these properties, we suggest a tuning procedure for the PI controller. We that the control algorithm in this paper, based on the linear system theory, is well work by simulation, and the LPD system modeling and control are more easy treatment for soccer robot.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.55-59
• /
• 1991

This paper presents a method for designing a full state feedback linear static control law. This will stabilize a given linear uncertain system and also guarantee the performance of the system. The uncertain systems are described by state equation which contains uncertain parameters in system and input distribution matrices. The method is based on the guaranteed cost control of Chang and Peng(1972). The controller gain can be obtained by the solution of a algebraic Riccati equation in which the input weighting matrices depend on the uncertainty bounds. The algebraic Riecati equation in this paper is same as that of weighted LQ regulator problem.

• 제어로봇시스템학회:학술대회논문집
• /
• 1988.10a
• /
• pp.639-642
• /
• 1988

In this paper, the problem of regulation in the presence of a known firing disturbance is considered. We show how one can apply a disturbance-utilizing control(DUC) theory to a actual gun servo model. Applied disturbance-utilizing control theory is established by combining LQ regulator and reduced order observer in the discrete time domain. To see the performance of the applied method, computer simulation results are given.