Object recall using an experience database to accelerate robot action planning

Richard Redpath; Jon Timmis; Martin Trefzer

doi:10.1109/IROS.2017.8205963

Object recall using an experience database to accelerate robot action planning

Richard Redpath, Jon Timmis, Martin Trefzer

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Robot interaction planning is a computationally expensive process which rarely makes use of previous experiences in a deliberative manner. This paper addresses this issue by examining dimensionality reduction techniques to allow comparison of objects in a robot's environment based on the way they react to robot manipulation. We compare a number of techniques which can map objects from an observation space, which may contain thousands of dimensions, to a lower dimensionality space - the embedding space - which allows objects to be compared in an efficient manner, making knowledge transfer between similar objects more computationally tractable.

Original language	English
Title of host publication	IROS 2017 - IEEE/RSJ International Conference on Intelligent Robots and Systems
Publisher	IEEE
Pages	1566-1571
Number of pages	6
Volume	2017-September
ISBN (Electronic)	9781538626825
DOIs	https://doi.org/10.1109/IROS.2017.8205963
Publication status	Published - 13 Dec 2017
Event	2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2017 - Vancouver, Canada Duration: 24 Sept 2017 → 28 Sept 2017

Conference

Conference	2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2017
Country/Territory	Canada
City	Vancouver
Period	24/09/17 → 28/09/17

Access to Document

10.1109/IROS.2017.8205963

Cite this

@inproceedings{0c267b1e042f4c7487690b5d45737216,

title = "Object recall using an experience database to accelerate robot action planning",

abstract = "Robot interaction planning is a computationally expensive process which rarely makes use of previous experiences in a deliberative manner. This paper addresses this issue by examining dimensionality reduction techniques to allow comparison of objects in a robot's environment based on the way they react to robot manipulation. We compare a number of techniques which can map objects from an observation space, which may contain thousands of dimensions, to a lower dimensionality space - the embedding space - which allows objects to be compared in an efficient manner, making knowledge transfer between similar objects more computationally tractable.",

author = "Richard Redpath and Jon Timmis and Martin Trefzer",

year = "2017",

month = dec,

day = "13",

doi = "10.1109/IROS.2017.8205963",

language = "English",

volume = "2017-September",

pages = "1566--1571",

booktitle = "IROS 2017 - IEEE/RSJ International Conference on Intelligent Robots and Systems",

publisher = "IEEE",

address = "United States",

note = "2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2017 ; Conference date: 24-09-2017 Through 28-09-2017",

}

Redpath, R , Timmis, J & Trefzer, M 2017, Object recall using an experience database to accelerate robot action planning. in IROS 2017 - IEEE/RSJ International Conference on Intelligent Robots and Systems. vol. 2017-September, 8205963, IEEE, pp. 1566-1571, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2017, Vancouver, Canada, 24/09/17. https://doi.org/10.1109/IROS.2017.8205963

TY - GEN

T1 - Object recall using an experience database to accelerate robot action planning

AU - Redpath, Richard

AU - Timmis, Jon

AU - Trefzer, Martin

PY - 2017/12/13

Y1 - 2017/12/13

N2 - Robot interaction planning is a computationally expensive process which rarely makes use of previous experiences in a deliberative manner. This paper addresses this issue by examining dimensionality reduction techniques to allow comparison of objects in a robot's environment based on the way they react to robot manipulation. We compare a number of techniques which can map objects from an observation space, which may contain thousands of dimensions, to a lower dimensionality space - the embedding space - which allows objects to be compared in an efficient manner, making knowledge transfer between similar objects more computationally tractable.

AB - Robot interaction planning is a computationally expensive process which rarely makes use of previous experiences in a deliberative manner. This paper addresses this issue by examining dimensionality reduction techniques to allow comparison of objects in a robot's environment based on the way they react to robot manipulation. We compare a number of techniques which can map objects from an observation space, which may contain thousands of dimensions, to a lower dimensionality space - the embedding space - which allows objects to be compared in an efficient manner, making knowledge transfer between similar objects more computationally tractable.

UR - http://www.scopus.com/inward/record.url?scp=85041957228&partnerID=8YFLogxK

U2 - 10.1109/IROS.2017.8205963

DO - 10.1109/IROS.2017.8205963

M3 - Conference contribution

AN - SCOPUS:85041957228

VL - 2017-September

SP - 1566

EP - 1571

BT - IROS 2017 - IEEE/RSJ International Conference on Intelligent Robots and Systems

PB - IEEE

T2 - 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2017

Y2 - 24 September 2017 through 28 September 2017

ER -

Object recall using an experience database to accelerate robot action planning

Abstract

Conference

Access to Document

Other files and links

Cite this