eprintid: 10083728 rev_number: 24 eprint_status: archive userid: 608 dir: disk0/10/08/37/28 datestamp: 2020-05-20 12:03:50 lastmod: 2021-11-23 23:43:18 status_changed: 2020-05-20 12:03:50 type: proceedings_section metadata_visibility: show creators_name: Kamthe, S creators_name: Peters, J creators_name: Deisenroth, MP title: Multi-modal filtering for non-linear estimation ispublished: pub divisions: UCL divisions: B04 divisions: C05 divisions: F48 keywords: Science & Technology, Technology, Acoustics, Engineering, Electrical & Electronic, Engineering, State estimation, Non-linear dynamical systems, Non-Gaussian filtering, Gaussian sum note: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. abstract: Multi-modal densities appear frequently in time series and practical applications. However, they are not well represented by common state estimators, such as the Extended Kalman Filter and the Unscented Kalman Filter, which additionally suffer from the fact that uncertainty is often not captured sufficiently well. This can result in incoherent and divergent tracking performance. In this paper, we address these issues by devising a non-linear filtering algorithm where densities are represented by Gaussian mixture models, whose parameters are estimated in closed form. The resulting method exhibits a superior performance on nonlinear benchmarks. date: 2014-07-14 date_type: published publisher: IEEE official_url: https://doi.org/10.1109/ICASSP.2014.6855154 oa_status: green full_text_type: other language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 1705330 doi: 10.1109/ICASSP.2014.6855154 publication_declined: 2020-08-13T22:37:23BST lyricists_name: Deisenroth, Marc lyricists_id: MDEIS71 actors_name: Deisenroth, Marc actors_id: MDEIS71 actors_role: owner full_text_status: public publication: 2014 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) place_of_pub: Florence, Italy pagerange: 7979-7983 pages: 5 event_title: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) event_location: Florence, ITALY issn: 1520-6149 book_title: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) citation: Kamthe, S; Peters, J; Deisenroth, MP; (2014) Multi-modal filtering for non-linear estimation. In: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). (pp. pp. 7979-7983). IEEE: Florence, Italy. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10083728/1/1401.0077v1.pdf