Robust adaptive neural control of flexible hypersonic flight vehicle with dead-zone input nonlinearity

Abstract

This paper presents adaptive dynamic surface control for the flexible model of hypersonic flight vehicle in the presence of unknown dynamics and input nonlinearity. By modeling the flexible coupling as disturbance of rigid body, based on the functional decomposition, the dynamics is divided into attitude subsystem and velocity subsystem. Flight path angle, pitch angle, and pitching rate are involved in the attitude subsystem. To eliminate the inherent problem of “explosion of complexity” in back-stepping, the dynamic surface control is investigated to construct the controller. Furthermore, direct neural control with robust design is proposed without estimating the control gain function and in this way the singularity problem could be avoided. In the last step of dynamic surface design, through the use of Nussbaum-type function, stable adaptive control is presented for the unknown dynamics with time- varying control gain function. The uniform ultimate boundedness stability of the closed-loop system is guaranteed. Simulation result shows the feasibility of the proposed method.