Intentional Motion On-line Learning and Prediction

Dizan Alejandro Vasquez Govea 1 Thierry Fraichard 1 Olivier Aycard 1 Christian Laugier 1
1 E-MOTION - Geometry and Probability for Motion and Action
GRAVIR - IMAG - Graphisme, Vision et Robotique, Inria Grenoble - Rhône-Alpes
Abstract : Motion prediction for objects which are able to decide their trajectory on the basis of a planning or decision process (e.g. humans and robots) is a challenging problem. Most existing approaches operate in two stages: a) learning, which consists in observing the environment in order to identify and model possible motion patterns or plans and b) prediction, which uses the learned plans in order to predict future motions. In existing techniques, learning is performed on-line, hence, it is impossible to rene the existing knowledge on the basis of the new observations obtained during the prediction phase. This paper proposes a novel learning approach which represents plans as Hidden Markov Models and is able to estimate the parameters and structure of those models in an incremental fashion by using the Growing Neural Gas algorithm. Our experiments demonstrate that the technique works in real-time, is able to operate concurrently with prediction and that the resulting model produces long-term predictions.
Type de document :
Communication dans un congrès
Proc. of the Int. Conf. on Field and Service Robotics, Jul 2005, Port Douglas, Australia. 2005
Liste complète des métadonnées

Littérature citée [20 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00182039
Contributeur : Christian Laugier <>
Soumis le : mercredi 24 octobre 2007 - 18:34:19
Dernière modification le : mercredi 11 avril 2018 - 01:54:03
Document(s) archivé(s) le : lundi 12 avril 2010 - 00:31:58

Fichier

fsr05.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00182039, version 1

Collections

INRIA | UGA | IMAG

Citation

Dizan Alejandro Vasquez Govea, Thierry Fraichard, Olivier Aycard, Christian Laugier. Intentional Motion On-line Learning and Prediction. Proc. of the Int. Conf. on Field and Service Robotics, Jul 2005, Port Douglas, Australia. 2005. 〈inria-00182039〉

Partager

Métriques

Consultations de la notice

350

Téléchargements de fichiers

221