Learning from demonstration for locally assistive mobility aids

Poon, J; Cui, Y; Valls Miro, J; Matsubara, T

Learning from demonstration for locally assistive mobility aids

Poon, J Cui, Y Valls Miro, J

Matsubara, T

Permalink

Publisher:: Springer Science and Business Media LLC
Publication Type:: Journal Article
Citation:: International Journal of Intelligent Robotics and Applications, 2019, 3, (3), pp. 255-268
Issue Date:: 2019-09-01

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Accepted Manuscript VersionAdobe PDF (491.12 kB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Poon, J
dc.contributor.author	Cui, Y
dc.contributor.author	Valls Miro, J https://orcid.org/0000-0002-0083-7797
dc.contributor.author	Matsubara, T
dc.date.accessioned	2020-05-03T21:32:47Z
dc.date.available	2020-09-30T19:05:35Z
dc.date.issued	2019-09-01
dc.identifier.citation	International Journal of Intelligent Robotics and Applications, 2019, 3, (3), pp. 255-268
dc.identifier.issn	2366-5971
dc.identifier.issn	2366-598X
dc.identifier.uri	http://hdl.handle.net/10453/140426
dc.description.abstract	© 2019, The Author(s). Active assistive systems for mobility aids are largely restricted to environments mapped a-priori, while passive assistance primarily provides collision mitigation and other hand-crafted behaviors in the platform’s immediate space. This paper presents a framework providing active short-term assistance, combining the freedom of location independence with the intelligence of active assistance. Demonstration data consisting of on-board sensor data and driving inputs is gathered from an able-bodied expert maneuvring the mobility aid around a generic interior setting, and used in constructing a probabilistic intention model built with Radial Basis Function Networks. This allows for short-term intention prediction relying only upon immediately available user input and on-board sensor data, to be coupled with real-time path generation based upon the same expert demonstration data via Dynamic Policy Programming, a stochastic optimal control method. Together these two elements provide a combined assistive mobility system, capable of operating in restrictive environments without the need for additional obstacle avoidance protocols. Experimental results in both simulation and on the University of Technology Sydney semi-autonomous wheelchair in settings not seen in training data show promise in assisting users of power mobility aids.
dc.language	en
dc.publisher	Springer Science and Business Media LLC
dc.relation.ispartof	International Journal of Intelligent Robotics and Applications
dc.relation.isbasedon	10.1007/s41315-019-00096-1
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	This is a post-peer-review, pre-copyedit version of an article published in International Journal of Intelligent Robotics and Applications. The final authenticated version is available online at: https://dx.doi.org/10.1007/s41315-019-00096-1.	en_US
dc.title	Learning from demonstration for locally assistive mobility aids
dc.type	Journal Article
utslib.citation.volume	3
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	open_access	*
dc.date.updated	2020-05-03T21:32:44Z
pubs.issue	3
pubs.publication-status	Published
pubs.volume	3
utslib.start-page	255
utslib.citation.issue	3

Abstract:

© 2019, The Author(s). Active assistive systems for mobility aids are largely restricted to environments mapped a-priori, while passive assistance primarily provides collision mitigation and other hand-crafted behaviors in the platform’s immediate space. This paper presents a framework providing active short-term assistance, combining the freedom of location independence with the intelligence of active assistance. Demonstration data consisting of on-board sensor data and driving inputs is gathered from an able-bodied expert maneuvring the mobility aid around a generic interior setting, and used in constructing a probabilistic intention model built with Radial Basis Function Networks. This allows for short-term intention prediction relying only upon immediately available user input and on-board sensor data, to be coupled with real-time path generation based upon the same expert demonstration data via Dynamic Policy Programming, a stochastic optimal control method. Together these two elements provide a combined assistive mobility system, capable of operating in restrictive environments without the need for additional obstacle avoidance protocols. Experimental results in both simulation and on the University of Technology Sydney semi-autonomous wheelchair in settings not seen in training data show promise in assisting users of power mobility aids.

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

http://hdl.handle.net/10453/140426