Conference item icon

Conference item

Goal-conditioned end-to-end visuomotor control for versatile skill primitives

Abstract:
Visuomotor control (VMC) is an effective means of achieving basic manipulation tasks such as pushing or pick- and-place from raw images. Conditioning VMC on desired goal states is a promising way of achieving versatile skill primitives. However, common conditioning schemes either rely on task-specific fine tuning - e.g. using one-shot imitation learning (IL) - or on sampling approaches using a forward model of scene dynamics i.e. model-predictive control (MPC), leaving deployability and planning horizon severely limited. In this paper we propose a conditioning scheme which avoids these pitfalls by learning the controller and its conditioning in an end-to-end manner. Our model predicts complex action sequences based directly on a dynamic image representation of the robot motion and the distance to a given target observation. In contrast to related works, this enables our approach to efficiently perform complex manipulation tasks from raw image observations without predefined control primitives or test time demonstrations. We report significant improvements in task success over representative MPC and IL baselines. We also demonstrate our model's generalisation capabilities in challenging, unseen tasks featuring visual noise, cluttered scenes and unseen object geometries.
Publication status:
Published
Peer review status:
Peer reviewed

Actions


Access Document


Publisher copy:
10.1109/ICRA48506.2021.9560752

Authors


More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Oxford college:
New College
Role:
Author
ORCID:
0000-0003-1374-2858


Publisher:
IEEE
Host title:
2021 IEEE International Conference on Robotics and Automation (ICRA)
Pages:
1319-1325
Publication date:
2021-10-18
Acceptance date:
2021-02-28
Event title:
2021 International Conference on Robotics and Automation (ICRA 2021)
Event location:
Xi'an, China
Event website:
https://www.ieee-ras.org/component/rseventspro/event/1920-icra-2021
Event start date:
2021-05-30
Event end date:
2021-06-05
DOI:
EISSN:
2577-087X
ISSN:
1050-4729
EISBN:
978-1-7281-9077-8
ISBN:
978-1-7281-9078-5


Language:
English
Keywords:
Pubs id:
1233315
Local pid:
pubs:1233315
Deposit date:
2022-01-26

Terms of use



Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP