Fault tolerant tracking control for Near Space Hypersonic Vehicle via neural network

In this paper, a fault tolerant control (FTC) strategy is investigated for Near Space Hypersonic Vehicle (NSHV) based on neural network and adaptive backstepping design. Firstly, a radial basis function (RBF) neural network (NN) is used to approximate the nonlinear dynamics, a neural network observer is constructed to estimate the unknown system fault, the adaptive on-line parameter-updating laws are derived, and the stability of the state error dynamic is guaranteed. Then an adaptive backstepping based fault tolerant controller is designed for the faulty system. The asymptotical stability of the closed-loop system and uniform boundedness of the state tracking error are proved according to Lyapunov theorem. Finally, simulation results on the NSHV attitude dynamics demonstrate the effectiveness of the proposed scheme.