Bounded-error identification for closed-loop systems

This paper presents a scheme for the identification of a system which operates in closed-loop and in the presence of bounded output disturbances. Two algorithms are proposed to solve this identification problem. The first algorithm is an Optimal Bounding Ellipsoid (OBE) type algorithm. This first algorithm is analyzed and sufficient conditions for stability and convergence are established. Relaxation of these conditions leads to a second identification algorithm. The implementation of that second algorithm is realized in an iterative scheme. A numerical example is provided to show the efficiency of the scheme.