Abstract
In this paper we reveal that the typical subspace identification algorithms use nonparsimonious model formulations, with extra terms in the model that appear to be non-causal. These terms are the causes for inflated variance in the estimates and partially responsible for the loss of closed-loop identifiability. We then propose a parallel parsimonious formulation of a new subspace identification algorithrn and demonstrate the effectiveness of the proposed algorithrn via simulation.
Original language | English |
---|---|
Pages (from-to) | 1591-1596 |
Number of pages | 6 |
Journal | IFAC Proceedings Volumes |
Volume | 36 |
Issue number | 16 |
DOIs | |
Publication status | Published - Sept 2003 |
Externally published | Yes |
Event | 13th IFAC Symposium on System Identification (SYSID 2003) - , Netherlands Duration: 27 Aug 2003 → 29 Aug 2003 |
Bibliographical note
Acknowledgments: Financial support from National Science Foundation under CTS-9985074 and a Faculty Research Assignment grant from University of Texas is gratefully acknowledged.Keywords
- parallel projections
- parsimonious formulation
- subspace identification