Published February 2010
| public
Journal Article
A convex optimization approach to filtering in jump linear systems with state dependent transitions
- Creators
- Capponi, Agostino
Abstract
We introduce a new methodology to construct a Gaussian mixture approximation to the true filter density in hybrid Markovian switching systems. We relax the assumption that the mode transition process is a Markov chain and allow it to depend on the actual and unobservable state of the system. The main feature of the method is that the Gaussian densities used in the approximation are selected as the solution of a convex programming problem which trades off sparsity of the solution with goodness of fit. A meaningful example shows that the proposed method can outperform the widely used interacting multiple model (IMM) filter and GPB2 in terms of accuracy at the expenses of an increase in computational time.
Additional Information
© 2009 Elsevier Ltd. Received 22 March 2008; revised 1 July 2009; accepted 4 November 2009. Available online 4 January 2010. The material in this paper was presented at 2009 American Control Conference (ACC2009), St. Louis, Missouri, USA, June 1012, 2009. This paper was recommended for publication in revised form by Associate Editor George Yin under the direction of Editor Ian R. Petersen.Additional details
- Eprint ID
- 17700
- DOI
- 10.1016/j.automatica.2009.11.011
- Resolver ID
- CaltechAUTHORS:20100308-145723629
- Created
-
2010-03-15Created from EPrint's datestamp field
- Updated
-
2021-11-08Created from EPrint's last_modified field