Design of a fractional-order neural network-based fixed-time sliding mode controller for chaotic satellite attitude control and synchronization

Purpose This study aims to propose a novel fractional-order neural network-based fixed-time sliding mode controller to address the challenges of chaotic satellite attitude control, synchronization and anti-synchronization. By integrating fractional-order dynamics and neural network-based estimation, the proposed approach enhances robustness and adaptability in stabilizing complex systems. Design/methodology/approach The proposed control scheme introduces a novel sliding surface that leverages fractional calculus to improve convergence speed and control flexibility. A neural network is used to adaptively estimate unknown error dynamics, compensating for uncertainties in the system. The fixed-time stability property guarantees convergence within a predefined time, independent of initial conditions. Theoretical stability proofs and numerical simulations are conducted to validate the effectiveness of the proposed controller. Findings The results demonstrate that the proposed controller effectively stabilizes chaotic satellite attitude dynamics and ensures reliable synchronization and anti-synchronization. The approach exhibits strong robustness against system uncertainties and external disturbances while achieving fixed-time convergence. Originality/value This work advances the field by integrating fractional-order dynamics with a neural network-based adaptive mechanism in a fixed-time sliding mode control framework. The proposed method enhances convergence properties and robustness in controlling chaotic satellite attitude systems, providing a novel and effective solution for complex nonlinear dynamics.