Utility- and Plan-based Action Selection based on Probabilis(4)

Abstract. This paper describes the AGILO RoboCuppers 1 the RoboCup team of the image understanding group (FG BV) at the Technische Universit?t München. With a team of four Pioneer I robots, all equipped with CCD camera and a single board computer, we’ve

onesideopen.Theplaysthemselvesspecifyhighlyreactive,conditional,andproperlysynchronizedbehaviorfortheindividualplayersoftheteam.

Thehigh-levelcontrollerisrealizedasastructuredreactivecontroller(SRC)[4]andimplementedinanextendedRPLplanlanguage[10].Thehigh-levelcontrollerworksasfollows.ItexecutesasadefaultstrategythesituatedactionselectionthatwehavedescribedinSec.3.2.Atthesametime,thecontrollercontinuallymonitorstheestimatedgamesituationinordertodetectopportunitiesformakingplays.Ifanop-portunityisdetected,thecontrollerdecidesbasedoncircumstancesincludingthescoreandtheestimatedsuccessprobabilityoftheintendedplaywhetherornottoperformtheplay.

Ourresearchgoalsinthedevelopmentofthehigh-levelcontrollerincludethedevel-opmentofacomputationalmodelforplan-basedcontrolinverydynamicmulti-robotapplicationsandfortheintegrationoflearningprocessesintotheplan-basedcontrol.References

1.ActivMediaRobotics.http://www.77cn.com.cn/robots/.

2.DeltaNetworkSoftwareGmbH.http://www.77cn.com.cn/,http://www.delta-net.de.

3.RadioLANInc.http://www.77cn.com.cn.

4.M.Beetz.Structuredreactivecontrollers.JournalofAutonomousAgentsandMulti-AgentSystems,4:25–55,2001.

5.S.BuckandM.Riedmiller.LearningSituationDependantSuccessRatesofActionsinaRoboCupScenario.InR.MizoguchiandJ.Slaney,editors,Paci cRimInternationalConferenceonArti cialIntelligence(PRICAI),number1886inLectureNotesinArti cialIntelligence,page809.Springer-Verlag,2000.

6.S.Buck,U.Weber,M.Beetz,andTh.Schmitt.Multirobotpathplanningfordynamicevironments:Acasestudy.In5thInternationalWorkshoponRoboCup(RobotWorldCupSoccerGamesandConferences),2001submitted.

7.WolfgangEcksteinandCarstenSteger.ArchitectureforComputerVisionApplicationDe-velopmentwithintheHORUSSystem.JournalofElectronicImaging,6(2):244–261,April1997.

8.R.HanekandT.Schmitt.Vision-basedlocalizationanddatafusioninasystemofcooper-atingmobilerobots.InInternationalConferenceonIntelligentRobotsandSystems(IROS),2000.

9.M.Klupsch.Object-OrientedRepresentationofTime-VaryingDataSequencesinMulti-agentSystems.InN.C.Callaos,editor,InternationalConferenceonInformationSystemsAnalysisandSynthesis(ISAS’98),pages833–839,Orlando,FL,USA,1998.InternationalInstituteofInformaticsandSystemics(IIIS).

10.D.McDermott.Areactiveplanlanguage.ResearchReportYALEU/DCS/RR-864,Yale

University,1991.

11.R.Quinlan.Inductionofdecisiontrees.InMachineLearning1(1),1986.

12.D.Reid.Analgorithmfortrackingmultipletargets.IEEETransactionsonAutomaticCon-

trol,24(6):843–854,1979.

13.T.Schmitt,R.Hanek,S.Buck,andM.Beetz.Cooperativeprobabilisticstateestimation

forvision-basedautonomoussoccerrobots.In5thInternationalWorkshoponRoboCup(RobotWorldCupSoccerGamesandConferences),volume5ofLectureNotesinComputerScience.Springer-Verlag,2001submitted.

百度搜索“77cn”或“免费范文网”即可找到本站免费阅读全部范文。收藏本站方便下次阅读，免费范文网，提供经典小说教育文库Utility- and Plan-based Action Selection based on Probabilis(4)在线全文阅读。