Control strategy for state and input observer design
Résumé
In this note, an alternative to classical methods for observer design is discussed, based on a dual control approach: it is indeed highlighted how an observer (at least approximate) can be obtained for a system by designing a control law for an auxiliary copy of this system, so that it tracks the system output. An important ingredient in this approach is the use of high-gain in the control design. The strategy is illustrated by various examples, including the case of unknown input reconstruction.