Output feedback control of hypersonic vehicles based on neural network and high gain observer

In this paper,the output feedback control problem for a genetic hypersonic vehicle is considered under the restriction that only the vehicle's velocity and altitude are measurable. High gain observers (HGO) are utilized to provide estimation signals for unmeasurable derivatives of the vehicle's velocity and altitude. Neural network based feedforward function is designed to compensate for model uncertainties. The proposed control design require less knowledge of the hypersonic vehicle's dynamic model. A comprehensive stability analysis of the closed loop system under the output feedback control is carried to prove that the proposed control law yields semiglobal uniformly ultimately bounded tracking while keeping all the closed loop signals bounded. Numerical simulation results are presented to validate the proposed control design.