室内环境下结合里程计的双目视觉SLAM研究

:100624710(2009)0420466206SLAM1,2,1,1(1.,710048;2.,710048):SLAM,SLAM,SIFT,;,,,,,:SLAM;;;SIFT;:TP24:AResearchonBinocularVisionSLAMwithOdometerinIndoorEnvironmentWANGXiao2hua1,2,FUWei2ping1,SULi1(1.FacultyofMechanicalandPrecisionInstrumentEngineering,XianUniversityofTechnology,Xian710048,China;2.CollegeofElectronicandInformation,XianPolytechnicUniversity,Xian710048,China)Abstract:WiththeaimofsolvingthelowpositioningaccuracyandlowrobustnessproblemsofvisionSLAMalgorithm,ExtendedKalmanFilter(EKF)methodbasedonbinocularvisionandodometerispro2posedinthispaper.FeaturepointcanbeobtainedbyextractingimagefeatureswithimprovedSIFTalgo2rithm,andthevisionfeaturemapisconstituted.SLAMiscompletedbyusingtheinformationofbinocularvisionandrobotpositionwithEKF.Thismethodcaneithersolvethemonocularvisioninaccuracyprob2lemoffeaturepointinformationobtainedbyspecialinitializationmethodoravoidtheenormouscomputa2tionbroughtaboutbybinocularvisionodometerusingimageinformationtorestoremovementaswellasthein2robustdisadvantagesofmotionestimation.Theresultsfromsimulationexperimentsindicatethatintheunknownindoorenvironments,thisalgorithmoperationisstable,andthepositioningaccuracyishigh.Keywords:SLAM;binocularvision;odometer;SIFT;ExtendedKalmanFilter(EKF):2009208224:(10872160):(19722),,,,,E2mail:[e-mail](SimultaneousLocalizationandMapping,SLAM)[1,2]SLAM,,[3],SLAM,SLAM[4]SLAM[5]SLAM,CCD,[6,7],SLAMSLAM[8,9],,,664JournalofXianUniversityofTechnology(2009)Vol.25No.4,,SLAM,;SIFT,,,,1SLAMSLAM,11Fig.1Systemframework,SIFT,,;,,,,(EKF),SIFT(),,,2SIFTSIFT,SIFT[10]:128,SIFT128,,,128,,,,,,SIFT,1)1,1281,11,;2)SIFT128,8,8,45,0;45,,,764:SLAM,,128,,2SIFT0.4,,98236,2b12801024(2a)50%,2SIFTFig.2SIFTmatchingresults3SLAM,3.1,,33Fig.3Robotmotionmodel:Xr(k+1)=f(Xr(k),u(k))+r(k)(1),Xr(k)Rnk,u(k),r(k),Q(k)u(k)=[Dk,k],:xr(k+1)yr(k+1)r(k+1)=xr(k)+Dkk[cos(r(k)+k)-cosr(k)]yr(k)+Dkk[sin(r(k)+k)-sinr(k)]r(k)+k+xr(k)yr(k)r(k)(2),xr(k)yr(k)r(k)k,Dkk,kk3.2,44Fig.4SensorobservationmodelZ(k)k:Z(k)=h(X(k))+(k)(3)h,H,(k),R(k):Z(k)=xr+xrLcosr-yrLsinryr+xrLsinr+yrLcosr+(k)(4)864(2009)254,(xrL,yrL),r3.3(EKF)Taylor,EKF,,:X(k)=Xr(k)XL(k)(5),Xr(k)=[xr,yr,r]T,XL(k)=[X1(k),X2(k),,Xn(k)]Tn51:ku(k)k+1ŽX(k+1)‰P(k+1):ŽX(k+1)=f(^X(k)+u(k))(6)‰P(k+1)=¨F(k)P(k)¨FT(k)+Q(k)(7),¨F(k)f(k)X(k)2:Y(k+1)3:Z(k+1)4:,(k)K(k+1)(k)=Y(k+1)-Z(k+1)(8)K(k+1)=‰P(k+1)HT(k+1)(H(k+1)‰P(k+1)HT(k+1)+R(k+1))-1(9)5:^X(k+1)=‰X(k+1)+K(k+1)[Y(k+1)-Z(k+1)](10)P(k+1)=(I-K(k+1)H(k+1))P(k)(11)4,,,EKF,5s,0.1m/sXr(0)=[-30,-30,0],SIFT,,ŽX(0);ŽX(1),,SIFT,Y(1),,Z(1),K(1)();,Y(1)Z(1)ŽX(1),^X(1),,EKF25SIFT3,,()EKF(),,,5SLAMFig.5SLAMexperimentalresults964:SLAM5,SIFT,;,,650,XY0.07m,[7]SLAM0.1m,SLAM,,76Fig.6Erroranalysis7SLAMFig.7SLAMexperimentalresultsindifferentconditions()EKF(),,7,,,;,,;,074(2009)2545SLAM,SIFT,,SLAM:[1]SmithR,SelfM,ChessemanP.Ontherepresentationandestimationofspatialuncertainty[J].InternationalJournalofRoboticsResarch,1986,5(4):56258.[2]Durrant2WhyteH,BaileyT.Simultaneouslocalizationandmapping.PartI[J].IEEERoboticsandAutomationMaga2zine,2006,13(3):992108.[3],(WangYao2nan,YuHong2shan).(Areviewofsimultaneouslocalizationandmapbuildingalgo2rithmsformobilerobotsinunknownenvironment)[J].(ControlTheory&Applications),2008,25(1):57264.[4]DavisonAJ.Real2timesimultaneouslocalicationandmap2pingwithasinglecamera[J].IEEETransactiononPat2ternAnalysisandMachineIntelligence,2002,24(7):8652880.[5]KimGH,KimJS,HongKS.Vision2BasedSimultaneousLocalizationandMappingwithTwoCameras:2005IEEE/RSJInternationalConferenceonIntelligentRobotsandSys2tems[C].Tokyo:IEEEPress,2005:340123405.[6]OrtegaJS,LemaireT,DevyM,etal.AMoninDelayedvsUndelayedLandmarkInitializationforBearingonlySLAM:ProceedingoftheIEEEInternationalOnferenceonRoboticsandAutomationWorkshoponSLAM[C].Japan:IEEEPress,2005:123.[7],,(WangPeng2lin,ShiShou2dong,HongXiao2wei).SLAM(ASLAMalgorithmbasedonmonocularvisionando2dometer)[J].(ComputerSimulation),2008,25(10):1722175.[8],,(WuGong2wei,ZhouWen2hui,GuWei2kang).(Dis2parityspacebasedbinocularvisualodometry)[J].(ChineseJournalofSensorsandActuator),2007,20(6):143221436.[9]DavisonAJ,NobuyukiK.3DSimultaneousLocalizationandMapBuildingUsingActiveVisionforaRobotMovingonUndulatingTerrain:ProceedingsoftheIEEEInterna2tionalConferenceonComputerVisionandRecognization[C].Hawail:IEEEPress,2001:3842391.[10]LoweD.Distinctiveimagefeaturesfromscale2invariantkeypoints[J].InternationalJournalofComputerVision,2004,60(2):912110.()174:SLAM