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 conferencePaper

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 languageEnglish (US)
Pages247-253
Number of pages7
StatePublished - Jan 1 1998
EventProceedings of the 1998 2nd International Conference on Autonomous Agents - Minneapolis, MN, USA
Duration: May 9 1998May 13 1998

Other

OtherProceedings of the 1998 2nd International Conference on Autonomous Agents
CityMinneapolis, MN, USA
Period5/9/985/13/98

    Fingerprint

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Schmill, M. D., Rosenstein, M. T., Cohen, P. R., & Utgoff, P. (1998). Learning what is relevant to the effects of actions for a mobile robot. 247-253. Paper presented at Proceedings of the 1998 2nd International Conference on Autonomous Agents, Minneapolis, MN, USA, .