Prescribed-time fault-tolerant control of a spacecraft with time-varying moments of inertia and input constraint

This paper presents a prescribed-time attitude control framework for spacecraft with a deployable panel, addressing key challenges such as input saturation, actuator faults, and time-varying inertia. Unlike conventional approaches that require precise knowledge of inertial dynamics, the proposed method relies only on known upper bounds of the inertia matrix and panel deployment rate, enabling robust performance in the presence of parametric uncertainty. The control strategy is built upon a command-filtered backstepping architecture, which eliminates the explosion of terms commonly encountered in traditional backstepping by approximating virtual control derivatives through first-order filter with time-varying gains. To compensate for the approximation errors introduced by the filters, auxiliary variables are incorporated into the controller. Furthermore, a compensation term inspired by the modified tracking error concept is embedded into these auxiliary variables to decouple the control input from the effects of actuator faults and input saturation. This structure improves fault tolerance and ensures stable performance even under constrained actuation. The prescribed-time stability of the closed-loop system is rigorously established using Lyapunov-based analysis. Simulation results confirm that the proposed controller drives tracking errors to zero within the user-defined time, even in the presence of time-varying moments of inertia, input saturation, and multiple faults. Sensitivity analyses reveal fundamental trade-offs between convergence time and actuator limitations, and demonstrate the controller’s robustness across a wide range of initial conditions. Comparative evaluations show that the proposed method outperforms existing strategies in tracking accuracy, fault resilience, and robustness under concurrent disturbances and constraints.