An Intelligent Sliding Mode Controller Based on Nominal Model of Robotic Manipulator

A novel global sliding model control (GSMC) control law based on backstepping method was proposed for tracking control of robot manipulators with uncertainties. This control scheme consisted of two combined controllers. One was the integral backstepping controller for nominal model method, the other was the compensatory sliding mode controller for practical plant. This control scheme integrated the theory of variable structure control and the nonlinear mapping of neural network. a Radial Basic Function Neural Network (RBFNN) was used to adaptively learn the unknown bounds of system uncertainties. The chattering of sliding mode could be eliminated effectively. The control law could guarantee convergence of trajectory tracking error as well as robustness for external disturbances, parameter and unparameter uncertainties. The simulation studies have verified the effectiveness of the proposed algorithm.