Predefined-Time Sliding Mode Control With Neural Network Observer for Hypersonic Morphing Vehicles
Guangbin Cai, Yiming Shang, Yongqiang Xiao, Tong Wu, Haitao Liu
Abstract
This study addresses the challenges posed by unknown disturbances arising from the morphing actions of hypersonic morphing vehicles (HMVs) and external perturbations encountered during operation. A predefined-time sliding mode control (PTSMC) strategy is proposed for the HMV. Initially, a nonlinear longitudinal dynamic model of the HMV is formulated to characterize its dynamic behavior. Subsequently, a predefined-time stability criterion is established based on Lyapunov stability theory, which serves as the theoretical foundation for the controller design. The proposed control framework integrates radial basis function neural network (RBFNN) disturbance observers to estimate and compensate for unknown disturbances in real time. Furthermore, rigorous theoretical analysis demonstrates that the closed-loop system achieves predefined-time stability. To validate the efficacy of the proposed methodology, comprehensive numerical simulations are conducted. These simulations employ Logistic functions to dynamically adjust predefined time parameters, thereby enhancing control precision and adaptability.