Search: id:"swepub:oai:DiVA.org:kth-306450" >
Meta Reinforcement ...
-
Arndt, KarolAalto Univ, Espoo, Finland.
(author)
Meta Reinforcement Learning for Sim-to-real Domain Adaptation
- Article/chapterEnglish2020
Publisher, publication year, extent ...
-
IEEE,2020
-
printrdacarrier
Numbers
-
LIBRIS-ID:oai:DiVA.org:kth-306450
-
https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-306450URI
-
https://doi.org/10.1109/ICRA40945.2020.9196540DOI
Supplementary language notes
-
Language:English
-
Summary in:English
Part of subdatabase
Classification
-
Subject category:ref swepub-contenttype
-
Subject category:kon swepub-publicationtype
Notes
-
QC 20211217conference ISBN 978-1-7281-7395-5
-
Modern reinforcement learning methods suffer from low sample efficiency and unsafe exploration, making it infeasible to train robotic policies entirely on real hardware. In this work, we propose to address the problem of sim-to-real domain transfer by using meta learning to train a policy that can adapt to a variety of dynamic conditions, and using a task-specific trajectory generation model to provide an action space that facilitates quick exploration. We evaluate the method by performing domain adaptation in simulation and analyzing the structure of the latent space during adaptation. We then deploy this policy on a KUKA LBR 4+ robot and evaluate its performance on a task of hitting a hockey puck to a target. Our method shows more consistent and stable domain adaptation than the baseline, resulting in better overall performance.
Subject headings and genre
Added entries (persons, corporate bodies, meetings, titles ...)
-
Hazara, MurtazaAalto Univ, Espoo, Finland.;Katholieke Univ Leuven, Dept Mech Engn, Leuven, Belgium.;Flanders Make, Robot Core Lab, Lommel, Belgium.
(author)
-
Ghadirzadeh, AliKTH,Robotik, perception och lärande, RPL,Aalto Univ, Espoo, Finland(Swepub:kth)u1hbw8ng
(author)
-
Kyrki, VilleAalto Univ, Espoo, Finland.
(author)
-
Aalto Univ, Espoo, Finland.Aalto Univ, Espoo, Finland.;Katholieke Univ Leuven, Dept Mech Engn, Leuven, Belgium.;Flanders Make, Robot Core Lab, Lommel, Belgium.
(creator_code:org_t)
Related titles
-
In:2020 IEEE International Conference On Robotics And Automation (ICRA): IEEE, s. 2725-2731
Internet link
To the university's database