• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.14-19
• /
• 1993

In this paper, we proposed an indirect learning and direct adaptive control schemes using neural networks, i.e., composite adaptive neural control, for a class of continuous nonlinear systems. With the indirect learning method, the neural network learns the nonlinear basis of the system inverse dynamics by a modified backpropagation learning rule. The basis spans the local vector space of inverse dynamics with the direct adaptation method when the indirect learning result is within a prescribed error tolerance, as such this method is closely related to the adaptive control methods. Also hash addressing technique, similar to the CMAC functional architecture, is introduced for partitioning network hidden nodes according to the system states, so global neuro control properties can be organized by the local ones. For uniform stability, the sliding mode control is introduced when the neural network has not sufficiently learned the system dynamics. With proper assumptions on the controlled system, global stability and tracking error convergence proof can be given. The performance of the proposed control scheme is demonstrated with the simulation results of a nonlinear system.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.470-473
• /
• 1996

This paper presents an iterative learning control scheme for industrial manipulators. Based upon the frequency-domain analysis, the input update law of the learning controller is given together with a sufficient condition for the convergence of the iterative process in the frequency domain. The proposed learning control scheme is structurally simple and computationally efficient since it is independent joint control depending only on locally measured variables and it does not involve the computation of complicated nonlinear manipulator dynamics. Moreover, it is capable of canceling the unmodeled dynamics of the manipulator without even the parametric model. Several important aspects of the learning scheme inherent in the frequency-domain design are discussed and the control performance is demonstrated through computer simulations.

• Proceedings of the KIEE Conference
• /
• 2001.11c
• /
• pp.11-14
• /
• 2001

원자력 발전소의 증기 발생기는 증기량과 급수량에 대한 비 최소위상 특성과 비선형성, 그리고 입력 제한 특성을 가지고 있다. 이러한 특성들은 증기 발생기의 효과적인 수위 제어에 어려움을 주고 있다. 본 논문에서는 게인 스케줄링 기법과 변형된 슬라이딩 모드 제어 기법을 이용한 원전 증기 발생기 제어기를 제안한다. 또한 앞먹임 구조를 가진 PI 제어기를 설계하여 저출력 영역에서 제안된 슬라이딩 모드 제어기와 성능을 비교한다. 모의 실험 결과 제안된 슬라이딩 모드 제어기가 최대 수위, 최소 수위, 그리고 안정화 시간 면에서 개선된 성능을 보였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.173.4-173
• /
• 2001

This paper presents a stable nonlinear adaptive control for AUV(Autonomous Underwater Vehicle) by using neural network. AUV's dynamics are highly nonlinear, and their hydrodynamic coefficients vary with different operational conditions. In this paper, the nonlinear uncertainties of the AUV's dynamics are approximated by using LPNN(Linearly parameterized Neural Network). The presented controller is consist of three parallel terms; linear feedback control, sliding mode control, and adaptive control(LPNN). Lyapunov theory is used to guarantee the stability of tracking errors and neural network´s weights errors. Numerical simulations for nonlinear control of the AUV show the effectiveness of the proposed techniques.

• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.7
• /
• pp.701-705
• /
• 2014

This article presents a comparison among several yaw and roll motion controllers for vehicle rollover prevention. In the previous research, yaw and roll motion controllers can be independently designed for rollover prevention. Following this idea, several yaw and roll motion controllers are designed and compared in terms of rollover prevention. For the yaw motion control, PID, LQR, SMC (Sliding Mode Control) and TDC (Time-Delay Control) are adopted. For the roll motion control, LQR, LQ SOF (Static Output Feedback) control, PID, and SMC are adopted. To compare the performance of each controller, simulation is performed on a vehicle simulation package, CarSim$^{(R)}$. From simulation, TDC and LQ SOF are the best for yaw and roll motion control, respectively.

• Journal of the Korean Society for Precision Engineering
• /
• v.9 no.3
• /
• pp.29-41
• /
• 1992

According to the recent increase of demands for multi-function and economics on hydraulic excavator, it is required that excavator should have simple operation, higher and operational efficiency, however the modeling of engine/pump system of excavator is not prescribed by the paper. So, in this paper the modeling of engine/pump system of excavator is suggested by identification method from step response and verified effectiveness of identification system by comparing with experimental results which was conducted using PID controller. To improve the problem of parameter variation and modeling error in the system, sliding mode control was introduced and new switching surface was designed. This control algorithm was applied to a hydraulic excavator by simulation, and its effectiveness was verified, and the results of variable structure system for the excavator system using a output component was compared with that of full state feedback when load disturbances and system paramenter variation exist.

• Journal of Mechanical Science and Technology
• /
• v.18 no.2
• /
• pp.203-210
• /
• 2004

Attitude control law design for spacecraft large angle maneuvers is investigated in this paper. The feedback linearization technique is applied to the design of a nonlinear tracking control law. The output function to be tracked is the quaternion attitude parameter. The designed control law turns out to be a combination of attitude and attitude rate tracking commands. The attitude-only output function, therefore, leads to a stable closed-loop system following the given reference trajectory. The principal advantage of the proposed method is that it is relatively easy to produce reference trajectories and associated controller.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1995.10a
• /
• pp.249-253
• /
• 1995

본 연구에서는 잠재적 응용성이 큰 ER유체와 압전필름을 액튜에이터로 하는 하이브리드형 지능구조물을 제안한 후 능동 진동제어를 실시하였다. 먼저 중공(hollow)의 샌드위치 형태 복합재료(glass/epoxy)보에 ER유체와 압전필름을 각각 삽입과 접착을 하여 하이브리드형 지능구조물을 제작하였다. 그리고 각 매체의 액튜에이팅 특성을 고려하여, ER유체 액튜에이터(ERFA)는 전장부하 함수로 도출되는 구조물의 주파수응답을 특징으로 하였고, 압전필름 액튜에이터(PFA)는 신경 슬라이딩 모드 제어기 (neuro sliding mode controller : NSC)를 적용하였다. 이 두 액튜에이터가 동시에 작동하는 능동 진동제어계를 실험적으로 구현한 후 과도응답과 강제 응답에 대한 진동제어 성능을 단일 액튜에이터 작동시와 비교 고찰하여 제시된 하이브리드 액튜에이팅의 효과를 입증하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.439-445
• /
• 1990

All models of dynamical systems invariably have some measure of uncertainties associated with some of their dynamics. The recent approaches to establish robustness of stabilizing feedback control against the possible uncertainties have a serious limitation, that is their applicability only to the systems that satisfy the matching conditions. Such conditions are rarely met in general applications. If a particular system satisfies the matching conditions, the addition of an actuator will destroy the satisfaction of such conditions. In this paper, we develop robust control algorithm for uncertain multivariable systems in which the matching conditions are not necessarily met. We empoly Lyapunov's second method to derive robust stabilizing controllers which guarantee asymptotic stability against prescribed uncertainties. The derivation consists of transforming the original uncertain system to controllable canonical form and constructing a constant switching surface by designing the closed-loop characteristics as a function of the uncertainties. Numerical examples are discussed as illustrations.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.7-12
• /
• 1990

In this paper, we propose a new manipulator control scheme based on the CMAG neural network. The proposed control consists of two components. The feedforward component is an output of trained CMAC neural network and the feedback component is a modified sliding mode control. The CMAC accepts the position, velocity and acceleration of manipulator as input and outputs two values for the controller : One is the nominal torque used for feedforward compensation(M1 network) and the other is the inertia matrix related information used for the feedback component(M2 network). Since the used control algorithm guarantees the robust trajectory tracking in spite of modeling errors, the CMAC mapping errors due to the memory limitation are little worth consideration.