Neural-Network-Based High-Order Sliding Mode Control via High-Order Fully Actuated System Approach

A RBF network based high-order sliding mode control (SMC) via high-order fully actuated (HOFA) system approaches is studied in this paper. Firstly, a high-order sliding surface based on HOFA system approaches is proposed such that the sliding mode dynamics is obtained. By using radial basis function (RBF) neural network, the unknown function can be approximated. Then, the adaptive RBF sliding mode controller is designed for HOFA systems with the unknown parameters and uncertain nonlinearity vectors such that global asymptotical stability is ensured. Compared with the existing HOFA systems approaches, the proposed method can be applied to deal with the unknown continuous function. Finally, an example shows the validity of proposed methods.