Fuzzy control rules extraction from perception-based information using computing with words

In this paper, a novel approach to extract fuzzy control rules from perception-based information using computing with words (CW) is firstly presented. The approach is based on applying fuzzy numbers and their arithmetic operations. The fuzzy Lyapunov synthesis, which gives a linguistic description on the plant and the control objective, is used to design a stable fuzzy controller. Secondly, that the traditional fuzzy-control rules can be derived systematically, rather than heuristically, and their stability can be guaranteed is demonstrated. Moreover, by introducing fuzzy numbers and their arithmetic operations, the "words" represented by fuzzy numbers could be efficiently manipulated for fuzzy controller design. The performance and applicability of the proposed method are illustrated through the practical implementation of fuzzy control of a pole-balancing vehicle.