A sliding-mode robust controller for permanent magnet spherical actuator using neural networks

A novel robust sliding-mode controller using neural networks (NNs) is given for trajectory tracking control of permanent magnet spherical actuator (PMSA) with external disturbance and system model errors. The controller is established including sliding-mode scheme and the neural networks. The radial basis function (RBF) neural networks are chosen to approximate the unknown model and uncertainty, as well as to reduce the chattering of sliding-mode controller. The weighting coefficients of the neural networks are trained online. The robust control is employed to overcome the approximation errors from NNs. All the adaptive tuning laws for the NNs are derived from Lyapunov stability theorem. The stability of the closed-loop system, as well as the convergence of the position tracking errors to zero, is guaranteed by the proposed controller. Simulation results are provided to validate the effectiveness of the proposed controller.