Neural network observer-based predefined-time attitude control for morphing hypersonic vehicles

Considering the fast-time varying external disturbance and model uncertainties caused by large-scale morphing and cross-domain flights, this study proposes a novel predefined-time sliding mode control method for the attitude control of morphing vehicles during the large-scale morphing phase. Firstly, a predefined-time neural network disturbance observer is established for approximating system error states. By combining the radial basis function (RBF) neural network technology with the disturbance observer technology, accurate observation results can be obtained in the presence of complex disturbances. Subsequently, the estimations of lumped disturbances are integrated into the control law, resulting in the development of a predefined-time sliding mode controller for morphing vehicles. Rigorously proof the stability of both the proposed controller and the disturbance observer are provided based on the Lyapunov function analysis, and it is proved that the proposed control method guarantees the predefined-time stability and the boundedness of the closed-loop system. Finally, simulations are performed for the attitude control of large-scale morphing vehicles. The simulation results demonstrate that the proposed method not only achieves predefined-time control but also effectively addresses fast time-varying model uncertainties and external disturbances.