Robust fuzzy-neural tracking control for robot manipulators

The paper presents a robust fuzzy-neural control scheme for an n-link robot manipulator with uncertainties. In the proposed scheme a fuzzy neural network is utilized to construct the control input by approximating the unknown nonlinearities of dynamic systems. The parameters of the fuzzy neural network approximator are modified using the recently proposed fuzzy-neural algorithm named Online Sequential Fuzzy Extreme Learning Machine (OS-Fuzzy-ELM), where the parameters of the membership functions characterizing the linguistic terms in the if-then rules are assigned by random values independent from the training data. Different from the original OS-Fuzzy-ELM algorithm, the consequent parameters of if-then rules are updated based on the stable laws derived based on Lyapunov stability theorem and Barbalat’s lemma so that the asymptotical stability of the system can be guaranteed. Also a sliding mode controller is incorporated to compensate for the modelling error of fuzzy neural network. Finally the proposed robust fuzzy-neural controller is applied to control a two-link robot manipulator and the simulation results verify the effectiveness of the proposed control scheme.