IMPLEMENTATION OF A LEARNING FUZZY CONTROLLER

This article describes our efforts at designing and implementing a pra ctical learning fuzzy controller using inexpensive hardware. The contr oller engages basic control concepts and system-independent learning r ules to enable it to adapt in real time to unknown plants even when it starts with a vacuous initial control policy. The controller remains dormant when the plant is operating satisfactorily, and autonomously i nitiates on-line adaptation in real time when adverse performance is o bserved. The Intel-8031-based hardware implementation is geared for ex tensibility, robustness, and fault tolerance. Limited plant-dependent information is incorporated to tailor the hardware to applications. Th e design produces learning rates exceeding 200 reinforcements per seco nd. The controller thus is able to learn to control unknown plants in real time even while it is controlling them. Physical experiments indi cate that the learning fuzzy controller can rapidly and effectively de al with variations in plant characteristics, compensate for wear and t ear, and handle disturbances and noise.