• Sorted by Date • Classified by Publication Type • Classified by Research Category •
Kurt Dresner and Peter Stone. Learning Policy Selection for Autonomous Intersection Management. In AAMAS 2007 Workshop on Adaptive and Learning Agents, pp. 34–39, Honolulu, Hawaii, USA, May 2007.
[PDF]79.9kB [gzipped postscript]49.5kB
Few aspects of modern life inflict as high a cost on society astraffic congestion and automobile accidents. Current work in AI andIntelligent Transportation Systems aims to replace human drivers withautonomous vehicles capable of safely and efficiently navigatingthrough the most hazardous city streets. Once such vehicles arecommon, interactions between multiple vehicles will be possible.Traffic lights and stop signs, which were designed for human drivers,may no longer be the best method for intersection controlPreviously, we made the case for a reservation-based intersectioncontrol mechanism designed for autonomous vehicles, but compatiblewith human drivers. Including human drivers allows incrementaldeployability as well as support for those who drive for pleasure, butmay result in significantly suboptimal performance, as human driversmay be present in dramatically varying proportions. In this paper, wedevelop a learning-based approach to determine which variant of thecontrol mechanism will be most effective under given conditions, andthen combine the resulting predictor with our multiagent intersectionmanagement mechanism, enabling it to determine when and how it shouldalter its configuration to best suit the current trafficconditions. Our extension is fully implemented and tested insimulation, and we provide experimental results demonstrating itsefficacy.
@InProceedings{2007alag-dresner,
author="Kurt Dresner and Peter Stone",
title="Learning Policy Selection for Autonomous Intersection Management",
booktitle="AAMAS 2007 Workshop on Adaptive and Learning Agents",
address="Honolulu, Hawaii, USA",
month="May", year="2007",
pages= "34--39",
abstract={
Few aspects of modern life inflict as high a cost on society as
traffic congestion and automobile accidents. Current work in AI and
Intelligent Transportation Systems aims to replace human drivers with
autonomous vehicles capable of safely and efficiently navigating
through the most hazardous city streets. Once such vehicles are
common, interactions between multiple vehicles will be possible.
Traffic lights and stop signs, which were designed for human drivers,
may no longer be the best method for intersection control
Previously, we made the case for a reservation-based intersection
control mechanism designed for autonomous vehicles, but compatible
with human drivers. Including human drivers allows incremental
deployability as well as support for those who drive for pleasure, but
may result in significantly suboptimal performance, as human drivers
may be present in dramatically varying proportions. In this paper, we
develop a learning-based approach to determine which variant of the
control mechanism will be most effective under given conditions, and
then combine the resulting predictor with our multiagent intersection
management mechanism, enabling it to determine when and how it should
alter its configuration to best suit the current traffic
conditions. Our extension is fully implemented and tested in
simulation, and we provide experimental results demonstrating its
efficacy.
},
bib2html_rescat = {Autonomous Intersection Management},
bib2html_pubtype = {Workshop},
bib2html_funding = {NSF},
}
Generated by bib2html (written by Patrick Riley ) on Wed Jul 02, 2008 11:31:21