Communication publiée dans un ouvrage (Colloques, congrès, conférences scientifiques et actes)
Learning to Grasp on the Moon from 3D Octree Observations with Deep Reinforcement Learning
ORSULA, Andrej; Bøgh, Simon; OLIVARES MENDEZ, Miguel Angel et al.
2022In Proceedings of 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Peer reviewed
 

Documents


Texte intégral
IROS22_1460.pdf
Postprint Auteur (6.24 MB)
Final submission
Télécharger

Tous les documents dans ORBilu sont protégés par une licence d'utilisation.

Envoyer vers



Détails



Mots-clés :
Space Robotics and Automation; Reinforcement Learning; Deep Learning in Grasping and Manipulation
Résumé :
[en] Extraterrestrial rovers with a general-purpose robotic arm have many potential applications in lunar and planetary exploration. Introducing autonomy into such systems is desirable for increasing the time that rovers can spend gathering scientific data and collecting samples. This work investigates the applicability of deep reinforcement learning for vision-based robotic grasping of objects on the Moon. A novel simulation environment with procedurally-generated datasets is created to train agents under challenging conditions in unstructured scenes with uneven terrain and harsh illumination. A model-free off-policy actor-critic algorithm is then employed for end-to-end learning of a policy that directly maps compact octree observations to continuous actions in Cartesian space. Experimental evaluation indicates that 3D data representations enable more effective learning of manipulation skills when compared to traditionally used image-based observations. Domain randomization improves the generalization of learned policies to novel scenes with previously unseen objects and different illumination conditions. To this end, we demonstrate zero-shot sim-to-real transfer by evaluating trained agents on a real robot in a Moon-analogue facility.
Disciplines :
Sciences informatiques
Ingénierie aérospatiale
Auteur, co-auteur :
ORSULA, Andrej  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > Space Robotics
Bøgh, Simon;  Aalborg University > Department of Materials and Production > Robotics and Automation
OLIVARES MENDEZ, Miguel Angel ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > Space Robotics
MARTINEZ LUNA, Carol  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > Space Robotics
Co-auteurs externes :
yes
Langue du document :
Anglais
Titre :
Learning to Grasp on the Moon from 3D Octree Observations with Deep Reinforcement Learning
Date de publication/diffusion :
23 octobre 2022
Nom de la manifestation :
2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Lieu de la manifestation :
Kyoto, Japon
Date de la manifestation :
23/10/2022 → 27/10/2022
Manifestation à portée :
International
Titre de l'ouvrage principal :
Proceedings of 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Peer reviewed :
Peer reviewed
Disponible sur ORBilu :
depuis le 22 août 2022

Statistiques


Nombre de vues
318 (dont 58 Unilu)
Nombre de téléchargements
162 (dont 10 Unilu)

citations Scopus®
 
12
citations Scopus®
sans auto-citations
6
citations OpenAlex
 
13

Bibliographie


Publications similaires



Contacter ORBilu