Training on the Job — Collecting Experience with Hierarchical Hybrid Automata

Abstract : We propose a novel approach to experience collection for autonomous service robots performing complex activities. This approach enables robots to collect data for many learning problems at a time, abstract it and transform it into information specific to the learning tasks and thereby speeding up the learning process. The approach is based on the concept of hierarchical hybrid automata, which are used as transparent and expressive representational mechanisms that allow for the specification of these experience related capabilities independent of the program itself. The suitability of the approach is demonstrated through experiments in which a robot doing household chore performs experience-based learning.
Type de document :
Communication dans un congrès
30th German Conference on Artificial Intelligence (KI-2007), Sep 2007, Osnabrück, Germany. Lecture Notes in Computer Science, 4667, pp.473-476, 2007, KI 2007: Advances in Artificial Intelligence 30th Annual German Conference on AI, KI 2007, Osnabrück, Germany, September 10-13, 2007. Proceedings. 〈10.1007/978-3-540-74565-5_43〉
Liste complète des métadonnées

Littérature citée [8 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01361444
Contributeur : Alexandra Kirsch <>
Soumis le : mercredi 7 septembre 2016 - 12:06:18
Dernière modification le : jeudi 8 septembre 2016 - 11:26:32
Document(s) archivé(s) le : jeudi 8 décembre 2016 - 12:39:04

Fichier

kirsch07training.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Alexandra Kirsch, Michael Beetz. Training on the Job — Collecting Experience with Hierarchical Hybrid Automata. 30th German Conference on Artificial Intelligence (KI-2007), Sep 2007, Osnabrück, Germany. Lecture Notes in Computer Science, 4667, pp.473-476, 2007, KI 2007: Advances in Artificial Intelligence 30th Annual German Conference on AI, KI 2007, Osnabrück, Germany, September 10-13, 2007. Proceedings. 〈10.1007/978-3-540-74565-5_43〉. 〈hal-01361444〉

Partager

Métriques

Consultations de la notice

22

Téléchargements de fichiers

37