Robust fuzzy control for robot manipulators

A robust fuzzy control is developed for robot manipulators to guarantee both global stability and performance. Robot dynamics under consideration may include large nonlinear uncertainties, such as nonlinear load variations and unmodelled dynamics. Fuzzy sets are chosen based on performance requirements and stability regions of the control system. For each fuzzy set, a sub-control is designed, based on nonlinear robust control design using Lyapunov's direct method; this is blended with others into a final fuzzy control. The resulting control provides not only robust and global stability, but also more accurate control performance than fuzzy controls obtained from constant sub-controls. The proposed design is applied to a robot trajectory control problem and compared with a standard nonlinear robust controller. The simulation results show that the proposed control is effective and yields superior tracking performance.