Unknown inputs, robust state observation for linear parameter varying systems

The problem of estimating the largest possible component of the state of a parameter dependent linear system in presence of unknown inputs is considered. Using a geometric approach and suitable defined invariant subspaces, that capture the concept of robustness with respect to parameter variation, a procedure for detecting the largest component of the state that may possibly be estimated together with a construction of the observer system and a technique for analyzing its asymptotic behavior is provided.