A Singularly Perturbed System Approach to Adaptive Neural Back-stepping Control Design of Hypersonic Vehicles

This paper presents the design of neural adaptive flight control systems for the longitudinal dynamics of hypersonic vehicle. By considering the coupling between thrust and pitch moment, the proposed control strategy is derived from the solutions of a series of fast dynamical equations, which are designed based on the back-stepping control and singularly perturbed system approach. The RBF neural networks are employed to approximate the unknown hypersonic dynamics. Simulation results are included to show the effectiveness of the neural adaptive control method.