Forgetting in Marginalized Particle Filtering and its Relation to Forward Smoothing
2011 (English)Report (Other academic)
The problem of degeneracy in marginalized particle ﬁltering is addressed. In particular, we note that the degeneracy is caused by loss of entropy of the posterior distribution and design maximum entropy estimates to prevent this. The main technique used in this report is known as forgetting. Itis shown that it can be used to suppress the problem with degeneracy, however, it is not a proper cure for the problem of stationary parameters. The problem of marginal-marginalized particle ﬁlter for suﬃcient statistics is also studied. The resulting algorithm is found to have remarkable similarities with the algorithm known as forward smoothing.
Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2011. , 21 p.
LiTH-ISY-R, ISSN 1400-3902 ; 3009
Rao-Blackwellized particle filtering, maximum entropy principle, Bayesian filtering
IdentifiersURN: urn:nbn:se:liu:diva-97959ISRN: LiTH-ISY-R-3009OAI: oai:DiVA.org:liu-97959DiVA: diva2:650800