Analytic Moment-based Gaussian Process Filtering

File Description SizeFormat 
icml2009_finalCorrected.pdfAccepted version343.09 kBAdobe PDFView/Open
Title: Analytic Moment-based Gaussian Process Filtering
Authors: Deisenroth, MP
Huber, MF
Hanebeck, UD
Item 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). copyright 2009.
Editors: Bouttou, L
Littman, ML
Issue Date: 15-Sep-2009
URI: http://hdl.handle.net/10044/1/11621
Publisher: Omnipress
Start Page: 225
End Page: 232
Journal / Book Title: Proceedings of the 26th International Conference on Machine Learning (ICML 2009)
Copyright Statement: © 2009 The Authors
Notes: timestamp: 2009.04.11
Place of Publication: Montreal, QC, Canada
Publisher URL: http://www.cs.mcgill.ca/%20icml2009/papers/344_corrected.pdf
Appears in Collections:Computing



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons