State Estimation for Linear Constrained Systems with Unknown Input

This paper deals with the filtering problem for linear discrete constrained dynamic systems with unknown input. The constraint matrix and constraint vector in this system are allowed to vary in the value and in the dimension. The original full state is separated into two parts, and the estimate of the state is reduced to find the optimal estimate of a singular system. The estimable condition is researched and a recursive estimator for the original full state is presented. A rigorous mathematical induction is given to compare the performance of our approach to that of the existing method without constraint. A numerical example is presented to demonstrate the effectiveness of the new method.