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
Fig.1.(a)Theodo(goalkeeper),Hugibert(attacker)and(b)whattheyperceiveoftheworldaroundthem.
port.APALcolorCCDcameraismountedontopoftherobotconsoleandlinkedtotheS-VHSinputofthevideocapturecard.Gain,shuttertime,andwhitebalanceofthecameraareadjustedmanually.CurrentlyRGB-16imagesarecapturedwitharesolu-tionof384172(bottom60%ofhalfPALresolution).Forbetterballguidancewe’vemountedasimpleconcave-shapedbarinfrontofeachrobot.Acustommadekickingdeviceenablestherobottokicktheballindirectionoftherobot’scurrentorientation.3FundamentalSoftwareConcepts
Thesoftwarearchitecture[9]ofoursystemisbasedonseveralindependentconcur-rentmodules.Themodulesareorganizedhierarchicallyintomain,intermediate,andbasicmodules.Themainmodulesareimage(sensor)analysis,informationfusion,ac-tionselection,pathplanning,androbotcontrol.Besidethemainmodulesthesystememploysauxiliarymodulesformonitoringtherobots’states.ThekeysoftwaremethodsemployedbytheAGILORoboCupperssoftwareare:
1.vision-basedcooperativestateestimationfordynamicenvironments,
2.synergeticcouplingofprogrammingandexperience-basedlearningformovementcontrolandsituatedactionselection,and
3.plan-basedcontrolofrobotteamsusingstructuredreactivecontrollers(SRCs).
3.1SelfLocalization,ObjectTrackingandDataFusionThevisionmoduleisakeypartofoursystem.Givenarawvideostream,themod-ulehastoestimatetheposeoftherobotandthepositionsoftheballandopponentrobots.Low/medium-levelimageprocessingoperationsareperformedwiththehelpoftheimageprocessinglibraryHALCON5.2.3(formerlyknownasHORUS[7]).
Wehavedevelopedaprobabilistic,vision-basedstateestimationmethodforindi-vidual,autonomousrobots.Thismethodenablesateamofmobilerobotstoestimatetheirjointpositionsinaknownenvironmentandtrackthepositionsofautonomouslymovingotherobjects.Allposesandpositionsofthestateestimationmodulecontainadescriptionofuncertainty,i.e.acovariancematrix.Thestateestimatorsofdifferentrobotscooperatetoincreasetheaccuracyandreliabilityoftheestimationprocess.Thiscooperationbetweentherobotsenablesthemtotracktemporarilyoccludedobjectsand