Convergence analysis of instrumental variable recursive subspace identification algorithms

Guillaume Mercère (1) , Marco Lovera (2)

1 LAII [Poitiers] - Laboratoire d'automatique et d'informatique industrielle [EA 1219]
2 Dipartimento di Elettronica e Informazione

The convergence properties of a recently developed recursive subspace identification algorithm of the MOESP class are investigated in this paper. The algorithm operates on the basis of an extended instrumental variable (EIV) version of the propagator method for signal subspace estimation. It is proved that, under weak conditions on the input signal and the identified system, the considered recursive subspace identification algorithm converges to a consistent estimate of the propagator and, by extension, of the state space system matrices

Mots clés: Subspace methods; Recursive algorithms; Identification algorithms; State space models; Tracking systems; Convergence analysis

Domaines: Automatique / Robotique

Guillaume Mercère, Marco Lovera. Convergence analysis of instrumental variable recursive subspace identification algorithms. 14th IFAC Symposium on System Identification, SYSID'2006, Mar 2006, Newcastle, Australia. pp.CDROM. ⟨hal-00098152⟩