• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.33-36
• /
• 2007

최근에 인공지능 분야에서는, 국내외적으로 강화학습(reinforcement learning)에 관한 관심이 크게 증폭되고 있다. 강화학습의 최근 경향을 살펴보면, 크게 가치함수를 직접 활용하는 방법(value function-based methods), 제어 전략에 대한 탐색을 활용하는 방법(policy search methods), 그리고 액터-크리틱 방법(actor-critic methods)의 세가지 방향으로 발전하고 있음을 알 수 있다. 본 논문에서는 이중 세 번째 부류인 액터-크리틱 방법 중 NAC(natural actor-critic) 기법의 한 종류인 RLS-NAC(recursive least-squares based natural actor-critic) 알고리즘을 다양한 트레이스 감쇠계수를 사용하여 연속제어입력(real-valued control inputs)으로 제어되는 Kimura의 기는 로봇에 대해 적용해보고, 그 성능을 기존의 SGA(stochastic gradient ascent) 알고리즘을 이용하여 학습한 경우와 비교해보도록 한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.505-510
• /
• 2005

The RPO(randomized policy optimizer) algorithm, which utilizes probabilistic policy for the action selection, is a recently developed tool in the area of reinforcement learning, and has been shown to be very successful in several application problems. In this paper, we propose a modified RPO algorithm, whose critic network is adapted via RLS(Recursive Least Square) algorithm. In order to illustrate the applicability of the modified RPO method, we applied the modified algorithm to Kimura's robot and observed very good performance. We also developed a MATLAB-based animation program, by which the effectiveness of the training algorithms on the acceleration or the robot movement were observed.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.63-66
• /
• 2004

The SGA(stochastic gradient ascent) algorithm is one of the most important tools in the area of reinforcement learning, and has been applied to a wide range of practical problems. In particular, this learning method was successfully applied by Kimura et a1. [1] to the control of a simple creeping robot which has finite number of control input choices. In this paper, we considered the application of the SGA algorithm to Kimura's robot control problem for the case that the control input is not confined to a finite set but can be chosen from a infinite subset of the real numbers. We also developed a MATLAB-based robot animation program, which showed the effectiveness of the training algorithms vividly.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.395-398
• /
• 2007

최근 인공지능 분야에서는 강화학습(Reinforcement Learning)에 대한 관심이 크게 증폭되고 있으며, 여러 관련 분야에 적용되고 있다. 본 논문에서는 강화학습 기법 중 액터-크리틱 계열에 속하는 RLS-NAC 알고리즘을 활용하여 Kimura의 기는 로봇의 이동을 다룰 때에 중요 파라미터의 결정을 위하여 meta-learning 기법을 활용하는 방안에 고려한다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.04a
• /
• pp.217-220
• /
• 2005

The RPO algorithm is a recently developed tool in the area of reinforcement Loaming, And it has been shown In be very successful in several application problems. In this paper, we consider a robot-control problem utilizing a modified RPO algorithm in which its critic network is adapted via RLS(Recursive Least Square) algorithm. We also developed a MATLAB-based animation program, by which the effectiveness of the training algorithms were observed.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.93.2-93
• /
• 2001

This paper presents the implementation of fuzzy PI gain scheduling controller (FPICGS) for controlling flexible joint robot arms with uncertainties from time-varying load. The term FPICGS is called based on a combination of fuzzy PI control scheme with a set of rule bases. Principle of design for a FPICGS is given along with the implementation of the designed computer aided control system. The experiment reveals an effectiveness of the proposed control scheme for flexible joint robot arms driven by a DC motorhooked with a spring which both parameters are completely unknown parameters ...

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.663-668
• /
• 1989

Control methods to achieve efficient and accurate deburring robots are proposed. For efficiency, cutting speed is controlled adoptively with the cutting load. For accuracy, it adopts repetitive control. Since usual repetitive control cannot afford dynamical speed changes, the proposed method controls in an interpolating manner using several waveforms stored in the controller. Successful experimental results axe shown.

• 제어로봇시스템학회:학술대회논문집
• /
• 1995.10a
• /
• pp.522-525
• /
• 1995

A manipulation of a multiple contacted object by a Rotational Base and Single-jointed Finger mechanism(RBSF mechanism) is discussed. The manipulation is characterized by multiple contacts on an object and large motions of the object with sliding contacts. The kinematics and dynamics allowing sliding at multiple contacts are explored. The conditions for manipulation of an object at multiple contacts by the RBSF mechanism, which cannot exert arbitrary contact forces because it has a fewer number of joints than is required for active control, is presented.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.559-564
• /
• 2003

Recently, particle filters have attracted attentions for nonlinear state estimation. They evaluate a posterior probability distribution of the state variable based on observations in simulation using so-called importance sampling. However, degeneracy phenomena in the importance weights deteriorate the filter performance. A new filter, Evolution Strategies Based Particle Filter, is proposed to circumvent this difficulty and to improve the performance. Numerical simulation results illustrate the applicability of the proposed idea.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10b
• /
• pp.1533-1538
• /
• 1991

A parameter space approach for robust control system design is developed by reducing several design specifications to sign definite conditions. It is shown that the gain and phase margin constraints for the parametric perturbed plant hold if and only if the four Kharitonov systems satisfy the margins. On pole location, it is shown that D-stability of convex combinations (1-t)p(s)+tq(s) can be determined by the coefficients corresponding to p(s) and q(s) based on the sign definite condition. We show a method of PI-type robust control system design as a useful example.