A novel model predictive observer for switched systems using dissipativity theory

This study suggests an online strategy for robust observer design in discrete-time switched systems, combining model predictive control and dissipativity theory. The proposed design aims to overcome unexpected variations, handle physical limitations, and address optimization challenges. The model predictive control approach is utilized to construct an online, optimal, and constrained observer, while dissipativity theory helps attenuate the impact of disturbances on the estimation error dynamics. Additionally, a persistent dwell-time structure is employed to relax the dwell-time restrictions compared to other observer design approaches. The coordinated design is formulated as an optimization problem with linear matrix inequalities, incorporating constraints to regulate the energy variation rate of switching moments and ensure the estimation error converges to zero. The effectiveness of the suggested design is demonstrated through numerical simulations, which validate its success in achieving the desired objectives.