Merger of rule based and variable structure controller

Abstract

The concept of fuzzy variable structure controller is developed. In this method the error state space around the sliding surface is represented using linguistic variables. A knowledge based controller which decides the contribution of equivalent control and attractability control component in the total control input is designed. The contribution is decided based on how far the current state is from the sliding surface and the origin of the error state space. The performance of the proposed controller is demonstrated for a second order linear time invariant system which has parameter uncertainty. The simulation results show that there is a considerable reduction in chattering and control energy.<>