Data-Driven Unknown-Input Observers and State Estimation

Unknown-input observers (UIOs) allow for estimation of the states an LTI system without knowledge all inputs. In this letter, we provide a novel data-driven UIO based on behavioral theory and result known as Fundamental Lemma proposed by Jan Willems coworkers. We give necessary sufficient conditions data collected from existence providing asymptotically converging state estimates, propose purely algorithm their computation. Even though focus UIOs, our results also apply to standard case completely As example, method distributed in DC microgrids.