Robust Adaptive Sliding Mode Neural Networks Control for Industrial Robot Manipulators

This paper proposes an original robust adaptive controller by using Radial Basis Function Neural networks (RBFNNs) for industrial robot manipulators (IRMs) in uncertain dynamical environments. This suggested control structure combines sliding mode technique, RBFNNs approximation and adaptive technique to improve the high accuracy of the tracking control. The proposed RBFNNs can deal the small problems successful because of its simple structure, faster training update laws and better approximation for the unknown dynamic of IRMs. All the parameters of the proposed control system are determined by Lyapunov stability theorem, and tuned online by an adaptive learning algorithm. Therefore, the stability, robustness and desired tracking performance of RBFNNs for IRMs are guaranteed. The simulations and experimental performed on a three-link IRMs are proposed in comparison with proportional integral differential (PID) and adaptive Fuzzy (AF) control to prove the robustness and efficiency of the RBFNNs.