Adaptive tracking control based on online LS-SVM identifier

The paper proposes an adaptive tracking control scheme for completely unknown nonlinear system based on online least squares support vector machines (LS-SVM) identifier. An adaptive neural network (ANN) inverse controller is developed and the controller parameters are adjusted by gradient information of the unknown nonlinear system model. As well as, considering of the parameter regulating process of ANN, a proportional-integral-derivative (PID) controller is combined to improve the control performance in initial stage and enhance the system stability. The simulation experimental results show that the adaptive neural network inverse control scheme based on online LS-SVM model and is effective. Moreover, the combined control scheme can improve effectively the control performance, especially in the initial control stage for completely unknown nonlinear system.