A Vector Field-Based Method for Human Action Representation and Recognition During Human-Robot Collaboration

Manitta, M; Jayasuriya, M; Liu, D

A Vector Field-Based Method for Human Action Representation and Recognition During Human-Robot Collaboration

Manitta, M Jayasuriya, M Liu, D

Permalink

Publisher:: IEEE
Publication Type:: Conference Proceeding
Citation:: 2023 IEEE 19th International Conference on Automation Science and Engineering (CASE), 2023, 2023-August, pp. 1-7
Issue Date:: 2023-09-28

Closed Access

	Filename	Description	Size
	A_Vector_Field-Based_Method_for_Human_Action_Representation_and_Recognition_During_Human-Robot.pdf	Published version	1.3 MB	Adobe PDF	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	Manitta, M
dc.contributor.author	Jayasuriya, M
dc.contributor.author	Liu, D https://orcid.org/0000-0002-1581-5582
dc.date	2023-08-26
dc.date.accessioned	2024-01-10T02:13:19Z
dc.date.available	2024-01-10T02:13:19Z
dc.date.issued	2023-09-28
dc.identifier.citation	2023 IEEE 19th International Conference on Automation Science and Engineering (CASE), 2023, 2023-August, pp. 1-7
dc.identifier.isbn	9798350320695
dc.identifier.issn	2161-8070
dc.identifier.issn	2161-8089
dc.identifier.uri	http://hdl.handle.net/10453/174192
dc.description.abstract	This paper presents a vector field based method for human action representation and recognition to query human hand motions in a human robot shared workspace A vector field of an action is fashioned to span the workspace domain with velocity vectors that evolve to encompass the dynamic trajectory of its motion as it traverses a shared workspace This method facilitates comparison between known and prevailing actions with wide variations in starting positions and trajectory motion enabling robust action recognition without an onerous training phase Two preliminary experiments are conducted to validate the proposed method The preliminary results show early and consistent action recognition with a success rate of up to 93 33 when tested in the presence of a robot
dc.language	en
dc.publisher	IEEE
dc.relation.ispartof	2023 IEEE 19th International Conference on Automation Science and Engineering (CASE)
dc.relation.ispartof	2023 IEEE 19th International Conference on Automation Science and Engineering
dc.relation.isbasedon	10.1109/case56687.2023.10260582
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	A Vector Field-Based Method for Human Action Representation and Recognition During Human-Robot Collaboration
dc.type	Conference Proceeding
utslib.citation.volume	2023-August
utslib.location.activity	Auckland, New Zealand
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Strength - RI - Robotics Institute
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Mechanical and Mechatronic Engineering
utslib.copyright.status	closed_access	*
dc.date.updated	2024-01-10T02:13:17Z
pubs.finish-date	2023-08-30
pubs.publication-status	Published
pubs.start-date	2023-08-26
pubs.volume	2023-August

Abstract:

This paper presents a vector field based method for human action representation and recognition to query human hand motions in a human robot shared workspace A vector field of an action is fashioned to span the workspace domain with velocity vectors that evolve to encompass the dynamic trajectory of its motion as it traverses a shared workspace This method facilitates comparison between known and prevailing actions with wide variations in starting positions and trajectory motion enabling robust action recognition without an onerous training phase Two preliminary experiments are conducted to validate the proposed method The preliminary results show early and consistent action recognition with a success rate of up to 93 33 when tested in the presence of a robot

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

http://hdl.handle.net/10453/174192