eprintid: 554
rev_number: 9
eprint_status: archive
userid: 7
dir: disk0/00/00/05/54
datestamp: 2011-07-27 09:04:05
lastmod: 2011-08-04 07:29:08
status_changed: 2011-07-27 09:04:05
type: book_section
metadata_visibility: show
contact_email: alberto.bemporad@imtlucca.it
item_issues_count: 0
creators_name: Bemporad, Alberto
creators_name: Garulli, Andrea
creators_name: Paoletti, Simone
creators_name: Vicino, Antonio
creators_id: alberto.bemporad@imtlucca.it
creators_id: 
creators_id: 
creators_id: 
title: Set membership identification of piecewise affine models
ispublished: pub
subjects: QA75
subjects: TJ
divisions: CSA
full_text_status: none
keywords: Nonlinear identification; hybrid system; bounded error; data classification; parameter estimation
abstract: This paper addresses the problem of identification of piecewise affine (PWA)models, which involves the joint estimations of both the parameters of the affine submodels and the partition of the PWA map from data. According to ideas from set-membership identification, the key approach is to characterize the model by its maximum allowed prediction error, which is used as a tuning knob for traning off between prediction accuracy and model complexity. At initialization, the proposed procedure for PWA identification exploits a technique per partitioning an infeasible system of linear inequalities into a (possibly minimum) number of feasible subsystems. This provides both an initial clustering of the datapoints and a guess of the number of required submodels, which therefore is not fixed a priori. A refinement procedure is then applied in order to improve both data classification and parameter estimation. The partition of the PWA maps is finally estimated by considering multicategory classification techniques.
date: 2003
date_type: published
publication: 13th IFAC Symposium on System Identification
publisher: Elsevier
pagerange: 1789-1794
event_dates: Rotterdam, The Netherlands
refereed: TRUE
isbn: 978-0-08-043709-5
book_title: System Identification 2003  : a proceeding volume from the 13th IFAC Symposium on System Identification
editors_name: Van Den Hof, Paul
editors_name: Wahlberg, Bo
editors_name: Weiland, Siep
related_url_url: http://www.elsevierdirect.com/product.jsp?isbn=0080437095
related_url_type: pub
citation:   Bemporad, Alberto and Garulli, Andrea and Paoletti, Simone and Vicino, Antonio  Set membership identification of piecewise affine models.   In:  System Identification 2003 : a proceeding volume from the 13th IFAC Symposium on System Identification.   Elsevier, pp. 1789-1794.  ISBN 978-0-08-043709-5      (2003)