Fixed-Time Composite Neuro-Adaptive Prescribed Performance Control of Spacecraft Considering Model Uncertainties and External Disturbances

This paper presents a prescribed performance attitude control strategy for spacecraft that handles modeling uncertainties and external disturbances. A novel performance function, based on fixed-time stability, is introduced, where the convergence time of the tracking errors to a neighborhood of zero is pre-assigned by the user, independent of the initial conditions. A backstepping control method is employed to design a state feedback controller with a finite-time differentiator incorporated to prevent term explosion during the backstepping process. A combination of a neural network and a disturbance observer is used to address modeling uncertainties and external disturbances alongside composite learning to enhance estimation accuracy. The stability of the proposed control system is rigorously validated through the Lyapunov method. Simulation results demonstrate that the control strategy successfully drives the spacecraft's state variable errors to the vicinity of zero within the specified time frame, even in the presence of uncertainties and disturbances.