An adaptive ULR fuzzy controller through reinforcement learning

摘要

This paper presents an adaptive fuzzy controller using unidirectional linear response (ULR) elements. The basic functions for a fuzzy controller including membership, minimum and defuzzification functions are realized by the ULR elements. Because the ULR element has a diode-like characteristics, it can be implemented by a diode-connected MOS transistor in current-mode implementations. The hardware implementation of the fuzzy controller using the ULR elements should also be very simple and straight-forward. In this paper, we also apply the ULR fuzzy controller to an inverted pendulum problem and demonstrate the effectiveness of the proposed ULR controller architecture and its learning capability through reinforcements.