Adaptive L-two-gain controller for flexible spacecraft attitude tracking

An adaptive L-two-gain controller is proposed for the attitude tracking of flexible spacecraft with external disturbances and input constraint. Neural networks are employed for the approximation of unknown system dynamics; an adaptive controller is developed to learn the undetermined parameters. Secondly, a robust controller is designed to achieve the L-two tracking performance with a desired disturbance-attenuation level. Finally, to treat the input constraint problem, the system is augmented with an auxiliary input signal error system. An improved adaptive L-two-gain controller is adopted to rapidly recover the unconstrained input signal when it goes beyond the input constraints. Simulations are carried out to study the effectiveness of the proposed control scheme; results show the theoretical and practical advantages of this approach.