Publications
• Sorted by Date • Classified by Publication Type • Classified by Research Category • Classified by Research Group • Classified by Author Last Name •
Freek Stulp and Michael Beetz. Optimized Execution of Action Chains Using Learned Performance Models of Abstract Actions. In Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence (IJCAI), 2005.
Download
Abstract
Many plan-based autonomous robot controllers generate chains ofabstract actions in order to achieve complex, dynamically changing,and possibly interacting goals. The execution of these action chainsoften results in robot behavior that shows abrupt transitions betweensubsequent actions, causing suboptimal performance. The resulting motionpatterns are so characteristic for robots that people imitating robotic behavior will do so by making abrupt movements betweenactions. In this paper we propose a novel computation model for the executionof abstract action chains. In this computation model a robot firstlearns situation-specific performance models of abstract actions. Itthen uses these models to automatically specialize the abstractactions for their execution in a given action chain. Thisspecialization results in refined chains that are optimized forperformance. As a side effect this behavior optimization also appearsto produce action chains with seamless transitions between actions.
BibTeX Entry
@InProceedings{stulp05optimized,
author = {Freek Stulp and Michael Beetz},
title = {Optimized Execution of Action Chains Using Learned Performance Models of Abstract Actions},
booktitle = {Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence (IJCAI)},
year = {2005},
bib2html_pubtype = {Refereed Conference Paper},
bib2html_rescat = {Plan-based Robot Control, Robot Learning},
bib2html_groups = {IAS,AGILO},
bib2html_funding = {AGILO},
bib2html_keywords = {},
abstract = {Many plan-based autonomous robot controllers generate chains of
abstract actions in order to achieve complex, dynamically changing,
and possibly interacting goals. The execution of these action chains
often results in robot behavior that shows abrupt transitions between
subsequent actions, causing suboptimal performance. The resulting motion
patterns are so characteristic for robots that people imitating robotic behavior will do so by making abrupt movements between
actions. In this paper we propose a novel computation model for the execution
of abstract action chains. In this computation model a robot first
learns situation-specific performance models of abstract actions. It
then uses these models to automatically specialize the abstract
actions for their execution in a given action chain. This
specialization results in refined chains that are optimized for
performance. As a side effect this behavior optimization also appears
to produce action chains with seamless transitions between actions.}
}
Generated by bib2html.pl (written by Patrick Riley ) on Thu Jul 03, 2008 11:12:06 UTC

















