Two-way fuzzy adaptive identification and control of a flexible-joint robot arm

Abstract

Abstract The objective in this paper is to apply our proposed two-way fuzzy adaptive system that makes use of intuitionistic fuzzy sets to the identification and model-based control of a flexible-joint robot arm. Uncertainty and inconsistency are modelled in the proposed system such as uncertainty is the width of the interval introduced by the independent assignment of membership and nonmembership functions of the intuitionistic fuzzy sets; and inconsistency is the violation of the consistency inequality in this assignment. We reduce uncertainty and inconsistency through a two phase training. The first phase is to reduce inconsistency introduced by the inconsistent assignment of membership and nonmembership functions. The resultant system is an almost consistent two-way fuzzy adaptive system. Thus, an evaluation of the degree of reduction of inconsistency is needed and is carried out at the end of this phase by forming the shadowed set patterns of the membership and nonmembership functions after first phase of training. The system is further trained for a second phase in order to reduce uncertainty. The system performance has shown that this second phase of training renders the system totally one-way fuzzy adaptive.