Design of a fuzzy learning controller and its application to robot manipulators

Fuzzy logic control schemes related by linguistic control rules have been gaining much attention as a method of intelligent control in recent years. However, a fuzzy controller is, in itself, a nonlinear controller having several parameters to be properly determined. In this paper, the authors propose a fuzzy learning controller which has an ability of generating and modifying the control rule-base, and tuning input/output gains for the iterative task. Through simulation studies on a linear time-invariant (LTI) single-input single-output (SISO) system, and a 2-link robot manipulator, the authors evaluate the performance of the proposed controller.<>