Survey of analytical IV estimates for errors-in-variables model identification

Stéphane Thil (1) , Marion Gilson (2)

1 ELIAUS - Electronique, Informatique, Automatique et Systèmes
2 CRAN - Centre de Recherche en Automatique de Nancy

This paper deals with discrete-time errors-in-variables model identification, where output and input data are both perturbed by additive noises. The goal of this paper is to show how instrumental variable techniques may handle this situation; more precisely, the focus is put on instrumental variable methods having closed-form solutions. Several methods using second- and high-order statistics are reviewed, and some improvements and new results are also introduced. The discussions are illustrated by means of a numerical simulation.

Domaines: Automatique / Robotique

Stéphane Thil, Marion Gilson. Survey of analytical IV estimates for errors-in-variables model identification. 18th IFAC World Congress, IFAC WC'2011, Aug 2011, Milan, Italy. pp.13098-13103. ⟨hal-00691860v2⟩