• 한국지능시스템학회논문지
• /
• 제8권3호
• /
• pp.9-15
• /
• 1998

본 논문은 뉴로제어 및 반복학습 제어기법에 기반한 미지의 비선형시스템의 적응학습제어 방법을 제안한다. 제안된 제어 시스템에서 반복학습제어기는 새로운 기준 궤적에 대해 시스템의 출력이 원하는 궤적으로 정확히 수렴하도록 하는 적응과 단기간 제어정보를 기억하는 기능을 수행한다. 상대차수만 알고 있는 미지 시스템에 대한 박복학습 법칙이 학습이득은 신경회로망을 이용하여 추정된다. 반복학습제어기에 의해 습득된 제어정보는 장기메모리에 기반한 앞먹임 뉴로제어기로 이전되어 누적기억됨으로써 과거에 겸험된 기준 궤적에 대해서는 신속하게 추종할 수 있도록 한다. 2자유도 매니퓰레이터에 적용하여 제안된 기법의 타당성을 검증한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.230-235
• /
• 1992

We propose a method of generating data to train a neural network controller. The data can be prepared directly by an iterative learning technique which repeatedly adjusts the control input to improve the tracking quality of the desired trajectory. Instead of storing control input data in memory as in iterative learning control, the neural network stores the mapping between the control input and the desired output. We apply this concept to the trajectory control of a two link robot manipulator with a feedforward neural network controller and a feedback linear controller. Simulation results show good generalization of the neural network controller.

• 대한전기학회논문지
• /
• 제39권10호
• /
• pp.1098-1107
• /
• 1990

Decentralized iterative learning control methods are presented for a class of large scale interconnected linear dynamic systems, in which iterative learning controller in each subsystem operates on its local subsystem exclusively with no exchange of information between subsystems. Suffcient conditions for convergence of the algorithms are given and numerical examples are illustrated to show the validity of the algorithms. In particular, the algorithms are useful for the systems having large uncertainty of inter-connected terms.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1483-1486
• /
• 2002

In this paper, a new scheme of iterative loaming control of a robot manipulator is presented. The proposed method uses a fuzzy sliding mode controller(FSMC), which is designed based on the similarity between the fuzzy logic control(FLC) and the sliding mode control(SMC), for the feedback. With this, the proposed method makes possible fDr fast iteration and has advantages that no linear approximation is used for the derivation of the learning law or in the stability proof Full proof of the convergence of the fuzzy sliding base learning scheme Is given.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 춘계학술대회논문집
• /
• pp.855-861
• /
• 2002

This paper demonstrates the reconstruction of input signals from only the measured signal for the simulation and endurance test of automobiles. The aim of this research is concerned with input signal reconstruction using various iterative teaming algorithm under the condition of system characteristics. From a linear to nonlinear systems which provides the output signals are estimated in this algorithm which is based on the frequency domain. Our concerns are that the algorithm can assure an acceptable stability and convergence compared to the ordinary iterative learning algorithm. As a practical application, a f car model with nonlinear damper system is used to verify the restoration of input signal especially with a modified iterative loaming algorithm.

• 제어로봇시스템학회논문지
• /
• 제2권3호
• /
• pp.165-173
• /
• 1996

It is not desirable to have too large control input in control systems, because there are usually a limitation for the input magnitude and cost for the input energy. Previous papers in the iterative learning control did not considered on these points. In this paper, an iterative learning control method is proposed for a class of nonlinear systems with consideration on input magnitude by adopting a concept of cost function consisting of the output error and the input magnitude in quadratic form. We proposed a new input update law with an input penalty function. If we choose a reasonable input penalty function, the two control objectives, good command following and small input energy, can be achieved. The characteristics of the proposed method are shown in the simulation examples.

• International Journal of Control, Automation, and Systems
• /
• 제6권2호
• /
• pp.194-203
• /
• 2008

Iterative learning control (ILC) is a simple and effective method for the control of systems that perform the same task repetitively. ILC algorithm uses the repetitiveness of the task to track the desired trajectory. In this paper, we propose a PID (proportional plus integral and derivative) type ILC update law for control discrete-time single input single-output (SISO) linear time-invariant (LTI) systems, performing repetitive tasks. In this approach, the input of controlled system in current cycle is modified by applying the PID strategy on the error achieved between the system output and the desired trajectory in a last previous iteration. The convergence of the presented scheme is analyzed and its convergence condition is obtained in terms of the PID coefficients. An optimal design method is proposed to determine the PID coefficients. It is also shown that under some given conditions, this optimal iterative learning controller can guarantee the monotonic convergence. An illustrative example is given to demonstrate the effectiveness of the proposed technique.

• International Journal of Control, Automation, and Systems
• /
• 제6권2호
• /
• pp.171-179
• /
• 2008

This paper addresses the synthesis of an iterative learning controller for a class of linear systems with norm-bounded parameter uncertainties. We take into account an iterative learning algorithm with current cycle feedback in order to achieve both robust convergence and robust stability. The synthesis problem of the developed iterative learning control (ILC) system is reformulated as the ${\gamma}$-suboptimal $H_{\infty}$ control problem via the linear fractional transformation (LFT). A sufficient convergence condition of the ILC system is presented in terms of linear matrix inequalities (LMIs). Furthermore, the ILC system with fast convergence rate is constructed using a convex optimization technique with LMI constraints. The simulation results demonstrate the effectiveness of the proposed method.

• Journal of Platform Technology
• /
• 제6권4호
• /
• pp.34-40
• /
• 2018

현재 인공신경망은 단일 작업에 대해선 뛰어난 성능을 보이나, 다른 종류의 작업을 학습하면 이전 학습 내용을 잊어버리는 단점이 있다. 이를 catastrophic forgetting이라고 한다. 인공신경망의 활용도를 높이긴 위해선 이 현상을 극복해야 한다. catastrophic forgetting을 극복하기 위한 여러 노력이 있다. 하지만 많은 노력이 있었음에도 완벽하게 catastrophic forgetting을 극복하지는 못하였다. 본 논문에서는 여러 노력 중 elastic weight consolidation(EWC)에 사용되는 핵심 개념을 이용하여, 순차적 반복학습을 제시한다. 인공신경망 학습에 많이 쓰이는 MNIST를 확장한 EMNIST 데이터 셋을 이용하여 catastrophic forgetting 현상을 재현하고 이를 순차적 반복학습을 통해 극복하는 실험을 진행하였으며, 그 결과 모든 작업에 대해서 학습이 가능하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1989년도 한국자동제어학술회의논문집; Seoul, Korea; 27-28 Oct. 1989
• /
• pp.1097-1112
• /
• 1989

An iterative learning control scheme is incorporated to the computed torque method as a means to enhance the accuracy and the flexibility. A learning rule is constructed by utilizing a gradient descent algorithm and data compressing techniques are illustrated. Computer simulation results show a good performance of the scheme under a relatively high speed and a heavy payload condition.