Learning Control for a Robotic Manipulator with Input Saturation

In this paper, adaptive neural-network control is designed for an n-DOF robotic manipulator system. In the tracking control design, both uncertainties and input saturation are considered. Stability of the closed-loop system is analyzed via the Lyapunov's direct method. The uncertain system is approximated by the radial basis function neural-networks (RBFNN) and the input saturation is solved by adding an auxiliary signal. Simulation studies are conducted to examine the effectiveness of the proposed model-based and RBFNN control.