Hierarchical and parameterized learning of pick-and-place manipulation from under-specified human demonstrations

Qian, K; Liu, H; Valls Miro, J; Jing, X; Zhou, B

Hierarchical and parameterized learning of pick-and-place manipulation from under-specified human demonstrations

Qian, K Liu, H Valls Miro, J

Jing, X Zhou, B

Permalink

Publisher:: Informa UK Limited
Publication Type:: Journal Article
Citation:: Advanced Robotics, 2020, 34, (13), pp. 858-872
Issue Date:: 2020-07-02

Closed Access

	Filename	Description	Size
	Hierarchical and parameterized learning of pick and place manipulation from under specified human demonstrations.pdf	Published version	4.33 MB		View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Qian, K
dc.contributor.author	Liu, H
dc.contributor.author	Valls Miro, J https://orcid.org/0000-0002-0083-7797
dc.contributor.author	Jing, X
dc.contributor.author	Zhou, B
dc.date.accessioned	2021-01-19T03:37:34Z
dc.date.available	2021-01-19T03:37:34Z
dc.date.issued	2020-07-02
dc.identifier.citation	Advanced Robotics, 2020, 34, (13), pp. 858-872
dc.identifier.issn	0169-1864
dc.identifier.issn	1568-5535
dc.identifier.uri	http://hdl.handle.net/10453/145480
dc.description.abstract	© 2020, © 2020 Informa UK Limited, trading as Taylor & Francis Group and The Robotics Society of Japan. Imitating manipulation skills through observing human demonstrations in everyday life is promising in allowing service robots to be programed quickly, as well as to perform human-like behaviors. Such a Learning by demonstration (LbD) problem is challenging because robots are expected to adapt their learned behaviors to the changes of task parameters and the environment, rather than simply cloning the human teacher's motion. In this paper, we propose a hierarchical and parameterized LbD framework that combines symbolic and trajectory learning of pick-and-place manipulation tasks. We have extended the two-step parameterized learning method with error compensation for learning Environment-adaptive Action Primitives (EaAPs), which is capable of adapting robot's reproduced trajectories to new task instances as well as environmental changes. To arrive at refined plans in situations of under-specified human demonstrations, we propose to model the semantics of demonstrated activities with PDDL-based skill scripts. Therefore, latent motion primitives that are impossible to be learned directly from observing human demonstration in noisy video data can be inferred. The proposed method is implemented as a hierarchical LbD framework and has been evaluated on real robot hardware to illustrate the effectiveness of the proposed approach.
dc.language	en
dc.publisher	Informa UK Limited
dc.relation.ispartof	Advanced Robotics
dc.relation.isbasedon	10.1080/01691864.2020.1778523
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0801 Artificial Intelligence and Image Processing, 0906 Electrical and Electronic Engineering, 1702 Cognitive Sciences
dc.subject.classification	Industrial Engineering & Automation
dc.title	Hierarchical and parameterized learning of pick-and-place manipulation from under-specified human demonstrations
dc.type	Journal Article
utslib.citation.volume	34
utslib.for	0801 Artificial Intelligence and Image Processing
utslib.for	0906 Electrical and Electronic Engineering
utslib.for	1702 Cognitive Sciences
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Strength - CAS - Centre for Autonomous Systems
pubs.organisational-group	/University of Technology Sydney
utslib.copyright.status	closed_access	*
dc.date.updated	2021-01-19T03:37:04Z
pubs.issue	13
pubs.publication-status	Published
pubs.volume	34
utslib.citation.issue	13

Abstract:

© 2020, © 2020 Informa UK Limited, trading as Taylor & Francis Group and The Robotics Society of Japan. Imitating manipulation skills through observing human demonstrations in everyday life is promising in allowing service robots to be programed quickly, as well as to perform human-like behaviors. Such a Learning by demonstration (LbD) problem is challenging because robots are expected to adapt their learned behaviors to the changes of task parameters and the environment, rather than simply cloning the human teacher's motion. In this paper, we propose a hierarchical and parameterized LbD framework that combines symbolic and trajectory learning of pick-and-place manipulation tasks. We have extended the two-step parameterized learning method with error compensation for learning Environment-adaptive Action Primitives (EaAPs), which is capable of adapting robot's reproduced trajectories to new task instances as well as environmental changes. To arrive at refined plans in situations of under-specified human demonstrations, we propose to model the semantics of demonstrated activities with PDDL-based skill scripts. Therefore, latent motion primitives that are impossible to be learned directly from observing human demonstration in noisy video data can be inferred. The proposed method is implemented as a hierarchical LbD framework and has been evaluated on real robot hardware to illustrate the effectiveness of the proposed approach.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/145480