Adaptive Non-Singular Fixed-Time Observer-Based Fault-Tolerant Attitude Tracking Control for Rigid Spacecraft with Reaction Wheels: Theory and Experimental Validation

Attitude tracking constitutes one of the most essential functions for rigid spacecraft operating in orbit. Nevertheless, the performance of spacecraft attitude control systems strongly depends on the operational integrity of onboard actuators, and actuator malfunctions can severely deteriorate tracking accuracy or even destabilize the entire system. Such adverse effects become particularly critical during transient maneuvers, where the spacecraft undergoes rapid rotational acceleration to achieve a prescribed attitude command. To cope with these demanding operating conditions, this work develops a fault-tolerant attitude tracking framework that combines a non-singular fixed-time controller with an adaptive fixed-time fault estimation mechanism. By treating actuator faults and external disturbances as a unified lumped uncertainty term, the proposed approach enhances system robustness under both nominal and faulty operating conditions. The adaptive estimation mechanism serves a crucial role in identifying the upper bound of the lumped uncertainty, thereby enabling reliable fault compensation and improving the overall fault-tolerant capability of the closed-loop system. In addition, rigorous stability proofs based on Lyapunov theory are provided to demonstrate that all closed-loop signals remain bounded and that the tracking errors converge within a fixed settling time independent of initial conditions. Furthermore, theoretical investigations reveal that the proposed non-singular fixed-time formulation achieves faster convergence characteristics compared with a class of existing fixed-time control strategies while simultaneously eliminating singularity-related issues. Finally, comprehensive experimental studies are carried out on a practical spacecraft test platform to verify the feasibility, robustness, and effectiveness of the proposed control methodology.