Direct Adaptive Control for a Class of Switched Nonlinear Systems Based on LS-SVM

A direct adaptive control scheme based on LS-SVM is proposed for a class of switched nonlinear dynamic systems with unknown nonlinearities in this paper. The LS-SVM technique is used to build the adaptive controller. The updating rule of LS-SVM parameters is derived from the switched Lyapunov function method, plus, an admissible switching signal with average dwell-time technique is given. The proposed control law can guarantee that the output tracking error and the states of the obtained closed-loop systems are asymptotically Lyapunov stable. A simulation example is given to validate the effectiveness and robustness of the proposed control approach.