• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제7권1호
• /
• pp.22-29
• /
• 2007

Two fuzzy controllers defined by 2 input variables cooperate to control a cartpole system in terms of balancing as well as centering. The cooperation is due to the BOXES scheme that selects one of the fuzzy controllers for each time step according to the content of box that is established through the critic of the control action by the fuzzy controllers. It is found that the control scheme is good at controlling the cartpole system so that the system is stabilized fast while the BOXES develops its ability to select proper fuzzy controller through experience.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제9권2호
• /
• pp.154-160
• /
• 2009

A control system composed of two fuzzy controllers is proposed to balance the pole as well as to move the cart to the center of the track of the cartpole system. The two fuzzy controllers are designed with 2 input variables respectively and their control characters are studied in order to devise a control scheme that alternates the two fuzzy controllers. It is found that the control system using the scheme works well even though there is some residual oscillations of the pole and the cart.

• 한국지능시스템학회논문지
• /
• 제16권3호
• /
• pp.349-356
• /
• 2006

운반차-막대 시스템을 제어하기 위하여 두 개의 2 차원 퍼지 제어기가 CMAC에 의해 협동하게 하는 제어 계략을 개발하였다. 제어계에서 한 제어기는 운반차의 변위와 속도, 다른 제어기는 막대의 각도와 각속도를 각각의 2 개의 입력 변수로 하고 운반차에 가하는 힘이 두 제어기의 출력 변수인데, 이 변수를 외부의 감독에 따라 CMAC이 학습하게 하여 협동 제어의 효과를 발휘한다. 제어계 구성과 CMAC 훈련에 의한 협동 계략의 단순함에 비하여, 제어계는 4 개의 입력 변수에 의한 퍼지 제어기나 다른 해석적 방법에 의한 것에 비해 손색없는 제어 성능을 보였다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제6권4호
• /
• pp.271-276
• /
• 2006

To implement a generalization of value functions in Adaptive Search Element (ASE)-reinforcement learning, CMAC (Cerebellar Model Articulation Controller) is integrated into ASE controller. ASE-reinforcement learning scheme is briefly studied to discuss how CMAC is integrated into ASE controller. Neighbourhood Sequential Training for CMAC is utilized to establish the look-up table and to produce discrete control outputs. In computer simulation, an ASE controller and a couple of ASE-CMAC neural network are trained to balance the inverted pendulum on a cart. The number of trials until the controllers are established and the learning performance of the controllers are evaluated to find that generalization ability of the CMAC improves the speed of the ASE-reinforcement learning enough to realize the cartpole control system.