Learning what is relevant to the effects of actions for a mobile robot

Matthew D. Schmill; Michael T. Rosenstein; Paul R. Cohen; Paul Utgoff

doi:10.1145/280765.280840

Learning what is relevant to the effects of actions for a mobile robot

Matthew D. Schmill, Michael T. Rosenstein, Paul R. Cohen, Paul Utgoff

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

3 Scopus citations

Abstract

We have developed a learning mechanism that allows robots to discover the conditional effects of their actions. Based on sensorimotor experience, this mechanism permits a robot to explore its environment and observe effects of its actions. These observations are used to learn a context operator difference table, a structure that relates circumstances (context) and actions (operators) to effects on the environment. From the context operator difference table, one can extract a relatively small set of state variables, which simplifies the problem of learning policies for complex activities. We demonstrate results with the Pioneer 1 mobile robot.

Original language	English (US)
Pages	247-253
Number of pages	7
DOIs	https://doi.org/10.1145/280765.280840
State	Published - 1998
Externally published	Yes
Event	Proceedings of the 1998 2nd International Conference on Autonomous Agents - Minneapolis, MN, USA Duration: May 9 1998 → May 13 1998

Other

Other	Proceedings of the 1998 2nd International Conference on Autonomous Agents
City	Minneapolis, MN, USA
Period	5/9/98 → 5/13/98

ASJC Scopus subject areas

General Engineering

Access to Document

10.1145/280765.280840

Cite this

@conference{f1685bb473314e9984505c5e5ce7e717,

title = "Learning what is relevant to the effects of actions for a mobile robot",

abstract = "We have developed a learning mechanism that allows robots to discover the conditional effects of their actions. Based on sensorimotor experience, this mechanism permits a robot to explore its environment and observe effects of its actions. These observations are used to learn a context operator difference table, a structure that relates circumstances (context) and actions (operators) to effects on the environment. From the context operator difference table, one can extract a relatively small set of state variables, which simplifies the problem of learning policies for complex activities. We demonstrate results with the Pioneer 1 mobile robot.",

author = "Schmill, {Matthew D.} and Rosenstein, {Michael T.} and Cohen, {Paul R.} and Paul Utgoff",

year = "1998",

doi = "10.1145/280765.280840",

language = "English (US)",

pages = "247--253",

note = "Proceedings of the 1998 2nd International Conference on Autonomous Agents ; Conference date: 09-05-1998 Through 13-05-1998",

}

TY - CONF

T1 - Learning what is relevant to the effects of actions for a mobile robot

AU - Schmill, Matthew D.

AU - Rosenstein, Michael T.

AU - Cohen, Paul R.

AU - Utgoff, Paul

PY - 1998

Y1 - 1998

N2 - We have developed a learning mechanism that allows robots to discover the conditional effects of their actions. Based on sensorimotor experience, this mechanism permits a robot to explore its environment and observe effects of its actions. These observations are used to learn a context operator difference table, a structure that relates circumstances (context) and actions (operators) to effects on the environment. From the context operator difference table, one can extract a relatively small set of state variables, which simplifies the problem of learning policies for complex activities. We demonstrate results with the Pioneer 1 mobile robot.

AB - We have developed a learning mechanism that allows robots to discover the conditional effects of their actions. Based on sensorimotor experience, this mechanism permits a robot to explore its environment and observe effects of its actions. These observations are used to learn a context operator difference table, a structure that relates circumstances (context) and actions (operators) to effects on the environment. From the context operator difference table, one can extract a relatively small set of state variables, which simplifies the problem of learning policies for complex activities. We demonstrate results with the Pioneer 1 mobile robot.

UR - http://www.scopus.com/inward/record.url?scp=0031681986&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031681986&partnerID=8YFLogxK

U2 - 10.1145/280765.280840

DO - 10.1145/280765.280840

M3 - Paper

AN - SCOPUS:0031681986

SP - 247

EP - 253

T2 - Proceedings of the 1998 2nd International Conference on Autonomous Agents

Y2 - 9 May 1998 through 13 May 1998

ER -

Learning what is relevant to the effects of actions for a mobile robot

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this