Advances in the Bayesian Occupancy Filter framework using robust motion detection technique for dynamic environment monitoring

Qadeer Baig 1, * Mathias Perrollaz 1, * Christian Laugier 1, *
* Auteur correspondant
1 E-MOTION - Geometry and Probability for Motion and Action
Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble
Abstract : The Bayesian Occupancy Filter provides a framework for grid-based monitoring of the dynamic environment. It allows to estimate dynamic grids, containing both information of occupancy and velocity. Clustering such grids then provides detection of the objects in the observed scene. In this paper we present recent improvements in this framework. First, multiple layers from a laser scanner are fused using opinion pool, to deal with conflicting information. Then a fast motion detection technique based on laser data and odometer/IMU information is used to separate the dynamic environment from the static one. This technique instead of performing a complete SLAM (Simultaneous Localization and Mapping) solution, is based on transferring occupancy information between consecutive data grids, the objective is to avoid false positives (static objects) like other DATMO approaches. Finally, we show the integration with Bayesian Occupancy Filter (BOF) and with the subsequent tracking module called Fast Clustering-Tracking Algorithm (FCTA). We especially show the improvements achieved in tracking results after this integration, for an intelligent vehicle application.
Type de document :
Article dans une revue
IEEE Robotics and Automation Magazine, Institute of Electrical and Electronics Engineers, 2014
Liste complète des métadonnées

Littérature citée [20 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00932691
Contributeur : Mathias Perrollaz <>
Soumis le : jeudi 17 juillet 2014 - 07:00:13
Dernière modification le : vendredi 12 octobre 2018 - 01:18:09
Document(s) archivé(s) le : lundi 17 novembre 2014 - 16:00:52

Fichier

IEEE_RAM_12-0143.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00932691, version 1

Collections

Citation

Qadeer Baig, Mathias Perrollaz, Christian Laugier. Advances in the Bayesian Occupancy Filter framework using robust motion detection technique for dynamic environment monitoring. IEEE Robotics and Automation Magazine, Institute of Electrical and Electronics Engineers, 2014. 〈hal-00932691〉

Partager

Métriques

Consultations de la notice

735

Téléchargements de fichiers

705