Persons Tracking with Gaussian Process Joint Particle Filtering

Persons Tracking with Gaussian Process Joint Particle Filtering. Jaakko Suutala Kaori Fujinami, J., R. In IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2010), 2010.
abstract bibtex

This paper presents an approach to tracking persons using Gaussian Processes (GP) and Particle Filtering (PF). We used a binary switch sensor floor, which provides a natural and transparent way to build an indoor positioning and tracking system. However, it poses many challenges by producing nonlinear non-Gaussian measurements of true location. To solve these issues we present a novel algorithm. It uses PF for Bayesian tracking and data association combined with learned GP regression to correct estimates. Furthermore, the proposed algorithm, called Gaussian Process Joint Particle Filtering (GPJPF), handles multiple targets, where each particle models the targets' states jointly. To handle the data association problem and interaction between targets in close proximity, a Markov Random Fields (MRF)-based motion model was applied. Along with the GP model, it can be used directly as an additional factor when calculating the importance weights of particles. In comparison, the proposed method outperforms conventional Gaussian process and particle filtering methods.

@inProceedings{
 title = {Persons Tracking with Gaussian Process Joint Particle Filtering},
 type = {inProceedings},
 year = {2010},
 issue = {160-165},
 city = {Kittilä, Finland},
 id = {92c748cb-3b9c-3443-bce6-96b9fe5e3b12},
 created = {2019-11-19T13:00:41.278Z},
 file_attached = {false},
 profile_id = {bddcf02d-403b-3b06-9def-6d15cc293e20},
 group_id = {17585b85-df99-3a34-98c2-c73e593397d7},
 last_modified = {2019-11-19T13:48:12.477Z},
 read = {false},
 starred = {false},
 authored = {false},
 confirmed = {true},
 hidden = {false},
 citation_key = {isg:1481},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {This paper presents an approach to tracking persons using Gaussian Processes (GP) and Particle Filtering (PF). We used a binary switch
sensor floor, which provides a natural and transparent way to build an indoor positioning and tracking system. However, it poses many
challenges by producing nonlinear non-Gaussian measurements of true location. To solve these issues we present a novel algorithm. It uses
PF for Bayesian tracking and data association combined with learned GP regression to correct estimates. Furthermore, the proposed algorithm,
called Gaussian Process Joint Particle Filtering (GPJPF), handles multiple targets, where each particle models the targets' states jointly. To handle the data association problem and interaction between targets in close proximity, a Markov Random Fields (MRF)-based motion model was applied. Along with the GP model, it can be
used directly as an additional factor when calculating the importance weights of particles. In comparison, the proposed method outperforms
conventional Gaussian process and particle filtering methods.},
 bibtype = {inProceedings},
 author = {Jaakko Suutala Kaori Fujinami, Juha Röning},
 booktitle = {IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2010)}
}

Downloads: 0

{"_id":"6B4W6wk7weaWugnRt","bibbaseid":"jaakkosuutalakaorifujinami-personstrackingwithgaussianprocessjointparticlefiltering-2010","authorIDs":[],"author_short":["Jaakko Suutala Kaori Fujinami, J., R."],"bibdata":{"title":"Persons Tracking with Gaussian Process Joint Particle Filtering","type":"inProceedings","year":"2010","issue":"160-165","city":"Kittilä, Finland","id":"92c748cb-3b9c-3443-bce6-96b9fe5e3b12","created":"2019-11-19T13:00:41.278Z","file_attached":false,"profile_id":"bddcf02d-403b-3b06-9def-6d15cc293e20","group_id":"17585b85-df99-3a34-98c2-c73e593397d7","last_modified":"2019-11-19T13:48:12.477Z","read":false,"starred":false,"authored":false,"confirmed":"true","hidden":false,"citation_key":"isg:1481","source_type":"inproceedings","private_publication":false,"abstract":"This paper presents an approach to tracking persons using Gaussian Processes (GP) and Particle Filtering (PF). We used a binary switch\nsensor floor, which provides a natural and transparent way to build an indoor positioning and tracking system. However, it poses many\nchallenges by producing nonlinear non-Gaussian measurements of true location. To solve these issues we present a novel algorithm. It uses\nPF for Bayesian tracking and data association combined with learned GP regression to correct estimates. Furthermore, the proposed algorithm,\ncalled Gaussian Process Joint Particle Filtering (GPJPF), handles multiple targets, where each particle models the targets' states jointly. To handle the data association problem and interaction between targets in close proximity, a Markov Random Fields (MRF)-based motion model was applied. Along with the GP model, it can be\nused directly as an additional factor when calculating the importance weights of particles. In comparison, the proposed method outperforms\nconventional Gaussian process and particle filtering methods.","bibtype":"inProceedings","author":"Jaakko Suutala Kaori Fujinami, Juha Röning","booktitle":"IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2010)","bibtex":"@inProceedings{\n title = {Persons Tracking with Gaussian Process Joint Particle Filtering},\n type = {inProceedings},\n year = {2010},\n issue = {160-165},\n city = {Kittilä, Finland},\n id = {92c748cb-3b9c-3443-bce6-96b9fe5e3b12},\n created = {2019-11-19T13:00:41.278Z},\n file_attached = {false},\n profile_id = {bddcf02d-403b-3b06-9def-6d15cc293e20},\n group_id = {17585b85-df99-3a34-98c2-c73e593397d7},\n last_modified = {2019-11-19T13:48:12.477Z},\n read = {false},\n starred = {false},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {isg:1481},\n source_type = {inproceedings},\n private_publication = {false},\n abstract = {This paper presents an approach to tracking persons using Gaussian Processes (GP) and Particle Filtering (PF). We used a binary switch\nsensor floor, which provides a natural and transparent way to build an indoor positioning and tracking system. However, it poses many\nchallenges by producing nonlinear non-Gaussian measurements of true location. To solve these issues we present a novel algorithm. It uses\nPF for Bayesian tracking and data association combined with learned GP regression to correct estimates. Furthermore, the proposed algorithm,\ncalled Gaussian Process Joint Particle Filtering (GPJPF), handles multiple targets, where each particle models the targets' states jointly. To handle the data association problem and interaction between targets in close proximity, a Markov Random Fields (MRF)-based motion model was applied. Along with the GP model, it can be\nused directly as an additional factor when calculating the importance weights of particles. In comparison, the proposed method outperforms\nconventional Gaussian process and particle filtering methods.},\n bibtype = {inProceedings},\n author = {Jaakko Suutala Kaori Fujinami, Juha Röning},\n booktitle = {IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2010)}\n}","author_short":["Jaakko Suutala Kaori Fujinami, J., R."],"bibbaseid":"jaakkosuutalakaorifujinami-personstrackingwithgaussianprocessjointparticlefiltering-2010","role":"author","urls":{},"downloads":0},"bibtype":"inProceedings","creationDate":"2019-11-19T13:17:06.052Z","downloads":0,"keywords":[],"search_terms":["persons","tracking","gaussian","process","joint","particle","filtering","jaakko suutala kaori fujinami"],"title":"Persons Tracking with Gaussian Process Joint Particle Filtering","year":2010}