Gaze allocation analysis for a visually guided manipulation task

Jose Nunez-Varela; Balaraman Ravindran; Jeremy L. Wyatt

doi:10.1007/978-3-642-33093-3_5

Gaze allocation analysis for a visually guided manipulation task

Jose Nunez-Varela^*, Balaraman Ravindran, Jeremy L. Wyatt

^*Corresponding author for this work

Computer Science

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

2 Citations (Scopus)

Abstract

Findings from eye movement research in humans have demonstrated that the task determines where to look. One hypothesis is that the purpose of looking is to reduce uncertainty about properties relevant to the task. Following this hypothesis, we define a model that poses the problem of where to look as one of maximising task performance by reducing task relevant uncertainty. We implement and test our model on a simulated humanoid robot which has to move objects from a table into containers. Our model outperforms and is more robust than two other baseline schemes in terms of task performance whilst varying three environmental conditions, reach/grasp sensitivity, observation noise and the camera's field of view.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages	44-53
Number of pages	10
Volume	7426 LNAI
DOIs	https://doi.org/10.1007/978-3-642-33093-3_5
Publication status	Published - 2012
Event	12th International Conference on Simulation of Adaptive Behavior, SAB 2012 - Odense, Denmark Duration: 27 Aug 2012 → 30 Aug 2012

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	7426 LNAI
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Conference

Conference	12th International Conference on Simulation of Adaptive Behavior, SAB 2012
Country/Territory	Denmark
City	Odense
Period	27/08/12 → 30/08/12

Keywords

decision making
Gaze control
reinforcement learning

ASJC Scopus subject areas

General Computer Science
Theoretical Computer Science

Access to Document

10.1007/978-3-642-33093-3_5

FP7_Collab_CogX: Cognitive Systems that Self-Understand and Self-Extend
Wyatt, J., Dearden, R. & Sloman, A.
European Commission, European Commission - Management Costs
1/05/08 → 30/06/12
Project: Research

Cite this

Nunez-Varela, J., Ravindran, B., & Wyatt, J. L. (2012). Gaze allocation analysis for a visually guided manipulation task. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7426 LNAI, pp. 44-53). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7426 LNAI). https://doi.org/10.1007/978-3-642-33093-3_5

Nunez-Varela, Jose ; Ravindran, Balaraman ; Wyatt, Jeremy L. / Gaze allocation analysis for a visually guided manipulation task. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7426 LNAI 2012. pp. 44-53 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ed1148299a4841a090f177d6c544aab4,

title = "Gaze allocation analysis for a visually guided manipulation task",

abstract = "Findings from eye movement research in humans have demonstrated that the task determines where to look. One hypothesis is that the purpose of looking is to reduce uncertainty about properties relevant to the task. Following this hypothesis, we define a model that poses the problem of where to look as one of maximising task performance by reducing task relevant uncertainty. We implement and test our model on a simulated humanoid robot which has to move objects from a table into containers. Our model outperforms and is more robust than two other baseline schemes in terms of task performance whilst varying three environmental conditions, reach/grasp sensitivity, observation noise and the camera's field of view.",

keywords = "decision making, Gaze control, reinforcement learning",

author = "Jose Nunez-Varela and Balaraman Ravindran and Wyatt, {Jeremy L.}",

year = "2012",

doi = "10.1007/978-3-642-33093-3_5",

language = "English",

isbn = "9783642330926",

volume = "7426 LNAI",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "44--53",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

note = "12th International Conference on Simulation of Adaptive Behavior, SAB 2012 ; Conference date: 27-08-2012 Through 30-08-2012",

}

Nunez-Varela, J, Ravindran, B & Wyatt, JL 2012, Gaze allocation analysis for a visually guided manipulation task. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 7426 LNAI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7426 LNAI, pp. 44-53, 12th International Conference on Simulation of Adaptive Behavior, SAB 2012, Odense, Denmark, 27/08/12. https://doi.org/10.1007/978-3-642-33093-3_5

Gaze allocation analysis for a visually guided manipulation task. / Nunez-Varela, Jose; Ravindran, Balaraman; Wyatt, Jeremy L.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7426 LNAI 2012. p. 44-53 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7426 LNAI).

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

TY - GEN

T1 - Gaze allocation analysis for a visually guided manipulation task

AU - Nunez-Varela, Jose

AU - Ravindran, Balaraman

AU - Wyatt, Jeremy L.

PY - 2012

Y1 - 2012

N2 - Findings from eye movement research in humans have demonstrated that the task determines where to look. One hypothesis is that the purpose of looking is to reduce uncertainty about properties relevant to the task. Following this hypothesis, we define a model that poses the problem of where to look as one of maximising task performance by reducing task relevant uncertainty. We implement and test our model on a simulated humanoid robot which has to move objects from a table into containers. Our model outperforms and is more robust than two other baseline schemes in terms of task performance whilst varying three environmental conditions, reach/grasp sensitivity, observation noise and the camera's field of view.

AB - Findings from eye movement research in humans have demonstrated that the task determines where to look. One hypothesis is that the purpose of looking is to reduce uncertainty about properties relevant to the task. Following this hypothesis, we define a model that poses the problem of where to look as one of maximising task performance by reducing task relevant uncertainty. We implement and test our model on a simulated humanoid robot which has to move objects from a table into containers. Our model outperforms and is more robust than two other baseline schemes in terms of task performance whilst varying three environmental conditions, reach/grasp sensitivity, observation noise and the camera's field of view.

KW - decision making

KW - Gaze control

KW - reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=84866010549&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-33093-3_5

DO - 10.1007/978-3-642-33093-3_5

M3 - Conference contribution

AN - SCOPUS:84866010549

SN - 9783642330926

VL - 7426 LNAI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 44

EP - 53

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

T2 - 12th International Conference on Simulation of Adaptive Behavior, SAB 2012

Y2 - 27 August 2012 through 30 August 2012

ER -

Nunez-Varela J, Ravindran B, Wyatt JL. Gaze allocation analysis for a visually guided manipulation task. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7426 LNAI. 2012. p. 44-53. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-33093-3_5

Gaze allocation analysis for a visually guided manipulation task

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Projects

FP7_Collab_CogX: Cognitive Systems that Self-Understand and Self-Extend

Cite this