Neural-adaptive chattering-free sliding mode attitude tracking control for spacecraft with fixed-time convergence

Precise spacecraft attitude tracking is essential for orbital operations in complex space environments, where lumped disturbances induce chattering phenomenon that degrades system stability. This paper investigates an adaptive chattering-free fixed-time sliding mode control (FxTSMC) scheme to tackle this issue. Firstly, a dynamic sliding surface is designed to guarantee fixed-time convergence of the attitude tracking error. To enhance robustness, a radial basis function neural network (RBFNN) with a fixed-time adaptive law is used to estimate the lumped disturbances characterized by model uncertainties and external disturbances in real time, without requiring prior knowledge. The proposed controller combines the RBFNN estimator with a fixed-time sliding mode (FxTSM) framework to improve tracking performance and reduce chattering effect. Theoretical analysis based on Lyapunov stability confirms the fixed-time convergence of both the system states and the neural network weights. Finally, simulation results manifest the effectiveness and priority of the proposed strategy in tracking accuracy and chattering reduction.