Predefined-time sliding mode control with RBF neural network for space tethered satellite deployment

AbstractThis paper presents a novel predefined-time sliding mode controller for the space tethered satellite (STS) system, incorporating a radial basis function neural network (RBFNN). First, a nonlinear dynamical model of the STS deployment is derived. Then, a modified nonsingular terminal sliding surface and corresponding nonsingular terminal sliding mode controller (NTSMC) with RBFNN are introduced. The predefined-time control ensures system convergence within a specified time, meeting mission requirements. Additionally, the RBFNN is employed to estimate and compensate for external disturbances, thus enhancing the system’s disturbance rejection capability. Meanwhile, the stability and predefined-time convergence of the proposed NTSMC are proved based on the Lyapunov theorem. Finally, numerical simulations of STS deployment are conducted, demonstrating the effectiveness and advantages of the proposed control scheme.