Attitude Control With Discrete-Time Prescribed Performance for Seeker Stabilized Platform via Event-Triggered Neural Approximation

This study focuses on a low-computational neural discrete-time attitude controller for the seeker stabilized platform, with the aim of ensuring fixed-time prescribed behaviors in tracking errors through an event-triggered approach. In contrast to existing reaching-law- based discrete-time prescribed performance control (DPPC) methodologies that rely on fully/partially accurate model plants, our approach establishes discrete-time protocols that accommodate unknown system dynamics while guaranteeing fixed convergence time and minimizing steady-state error to its utmost extent. This is achieved through a newly constructed design framework that completely deviates from current DPPC practices. Our framework incorporates fixed-time performance functions, the back-stepping procedure, and innovative accumulation terms, all of which synergistically enhance steady-state accuracy while obviating the need to calculate virtual controller differences. Furthermore, we utilize neural networks to approximate unknown system dynamics and devise an event-triggered mechanism instead of a time-based one for intermittently updating neural approximators, thereby effectively reducing computational costs. Finally, comparative validation results demonstrate that the proposed control precision has been enhanced by 27% to one or more orders of magnitude and the simulation run time has been reduced by more than 26% under identical conditions compared with existing strategies.