RBF Neural Network-Based Cooperative Electromagnetic Takeover Control for Large-Scale Failed Spacecraft
Chuang Liu, Jiayi Xu, Xiaokui Yue
Abstract
To achieve precise and smooth docking, this paper proposes a sliding mode control strategy for cooperative electromagnetic takeover involving multiple servicing spacecraft. First, a relative motion dynamics model for electromagnetic takeover between servicing and failed spacecraft is established based on a far-field electromagnetic force model. Next, an intelligent sliding mode controller is developed based on a radial basis function (RBF) neural network. The controller ensures fixed-time convergence and prescribed-time stability, with system stability rigorously proven using Lyapunov theory. Finally, an appropriate communication topology and control parameters are selected to conduct numerical simulations of multi-servicing spacecraft for cooperative electromagnetic takeover. Simulation results demonstrate that the proposed RBF neural network-based sliding mode control approach enables high-precision and smooth electromagnetic takeover under strong disturbance conditions, while maintaining the stability of the system's center of mass throughout the takeover process.