Analytic moment-based Gaussian process filtering.
File(s)
Published version
Author(s)
Deisenroth, Marc Peter
Huber, Marco F
Hanebeck, Uwe D
Type
Conference Paper
Abstract
We propose an analytic moment-based filter for nonlinear stochastic dynamic systems modeled by Gaussian processes. Exact expressions for the expected value and the covariance matrix are provided for both the prediction step and the filter step, where an additional Gaussian assumption is exploited in the latter case. Our filter does not require further approximations. In particular, it avoids finite-sample approximations. We compare the filter to a variety of Gaussian filters, that is, the EKF, the UKF, and the recent GP-UKF proposed by Ko et al. (2007).
Editor(s)
Danyluk, Andrea Pohoreckyj
Bottou, Léon
Littman, Michael L
Date Issued
2009
Citation
ICML '09 Proceedings of the 26th Annual International Conference on Machine Learning, 2009, pp.225-232
ISBN
978-1-60558-516-1
Publisher
ACM
Start Page
225
End Page
232
Journal / Book Title
ICML '09 Proceedings of the 26th Annual International Conference on Machine Learning
Copyright Statement
© 2009 by the author(s) / owner(s). The definitive Version of Record was published in IMCL '09 Proceedings, http://dx.doi.org/10.1145/1553374.1553403
Description
19.09.13 KB. Ok to add published version to spiral, authors are copyright holders.
Source
ICML '09
Place of Publication
Montreal, QC, Canada
Start Date
2009-06-14
Finish Date
2009-06-18
Coverage Spatial
Montreal, Canada