Wang Zhijie,Fang Jian'an,Shao ShihuangDeptartment of Automation and Electrical Information Engineering,China Textile University,Shanghai 200051
Abstract:
A new reinforcement method for learning fuzzy logic controllers is proposed. The reinforcement learning scheme is composed of two fuzzy logic rule bases: one acts as an adaptive critic, the other serves as a controller . The proposed method is tested on the cart - pole system . Simulation results show that the method has better learning performance than Anderson's neural network -based method.