Reinforcement learning for a vision based mobile robot

C. Gaskett; L. Fletcher; A. Zelinsky

Reinforcement learning for a vision based mobile robot

C. Gaskett^*, L. Fletcher, A. Zelinsky

^*Corresponding author for this work

Research output: Contribution to conference › Paper › peer-review

32 Citations (Scopus)

Abstract

Reinforcement learning systems improve behaviour based on scalar rewards from a critic. In this work vision based behaviours, servoing and wandering, are learned through a Q-learning method which handles continuous states and actions. There is no requirement for camera calibration, an actuator model, or a knowledgeable teacher. Learning through observing the actions of other behaviours improves learning speed. Experiments were performed on a mobile robot using a real-time vision system.

Original language	English
Pages	403-409
Number of pages	7
Publication status	Published - 2000
Event	2000 IEEE/RSJ International Conference on Intelligent Robots and Systems - Takamatsu, Japan Duration: 31 Oct 2000 → 5 Nov 2000

Conference

Conference	2000 IEEE/RSJ International Conference on Intelligent Robots and Systems
Country/Territory	Japan
City	Takamatsu
Period	31/10/00 → 5/11/00

Cite this

@conference{5e11068ea5ae49ea872355a3b120523c,

title = "Reinforcement learning for a vision based mobile robot",

abstract = "Reinforcement learning systems improve behaviour based on scalar rewards from a critic. In this work vision based behaviours, servoing and wandering, are learned through a Q-learning method which handles continuous states and actions. There is no requirement for camera calibration, an actuator model, or a knowledgeable teacher. Learning through observing the actions of other behaviours improves learning speed. Experiments were performed on a mobile robot using a real-time vision system.",

author = "C. Gaskett and L. Fletcher and A. Zelinsky",

year = "2000",

language = "English",

pages = "403--409",

note = "2000 IEEE/RSJ International Conference on Intelligent Robots and Systems ; Conference date: 31-10-2000 Through 05-11-2000",

}

TY - CONF

T1 - Reinforcement learning for a vision based mobile robot

AU - Gaskett, C.

AU - Fletcher, L.

AU - Zelinsky, A.

PY - 2000

Y1 - 2000

N2 - Reinforcement learning systems improve behaviour based on scalar rewards from a critic. In this work vision based behaviours, servoing and wandering, are learned through a Q-learning method which handles continuous states and actions. There is no requirement for camera calibration, an actuator model, or a knowledgeable teacher. Learning through observing the actions of other behaviours improves learning speed. Experiments were performed on a mobile robot using a real-time vision system.

AB - Reinforcement learning systems improve behaviour based on scalar rewards from a critic. In this work vision based behaviours, servoing and wandering, are learned through a Q-learning method which handles continuous states and actions. There is no requirement for camera calibration, an actuator model, or a knowledgeable teacher. Learning through observing the actions of other behaviours improves learning speed. Experiments were performed on a mobile robot using a real-time vision system.

UR - http://www.scopus.com/inward/record.url?scp=0034446356&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0034446356

SP - 403

EP - 409

T2 - 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems

Y2 - 31 October 2000 through 5 November 2000

ER -

Reinforcement learning for a vision based mobile robot

Abstract

Conference

Other files and links

Fingerprint

Cite this