Fuzzy Brain Emotional Learning Control System Design for Nonlinear Systems

Abstract

This paper aims to develop a new fuzzy neural network by incorporating a brain emotional learning controller (BELC) with the fuzzy inference rules. The brain has an orbitofrontal cortex and an amygdala. The BELC is a mathematical model, which imitates judgment and emotion of a brain. So that BELC contains two neural networks; the former is a sensory neural network and the latter is an emotional neural network. These two systems affect each other and this will effectively improve the approximation ability of BELC. By incorporating fuzzy inference system into the BELC, a novel fuzzy brain emotional learning controller (FBELC) is proposed which includes two fuzzy inference systems. The parameter updating rules have an interacting term between two systems, which will increase the learning performance of this system. Then, the developed FBELC is applied to control nonlinear systems. Two examples, a chaotic system and an inverted double pendulum system, are demonstrated to illustrate the effectiveness of the proposed control method. A comparison between the proposed FBELC with other controllers shows that the proposed controller can achieve better control performance than the other controllers.