動(dòng)態(tài)環(huán)境基于粒子濾波的移動(dòng)機(jī)器人自定位.pdf

1、浙江大學(xué)信息科學(xué)與工程學(xué)院碩士學(xué)位論文動(dòng)態(tài)環(huán)境基于粒子濾波的移動(dòng)機(jī)器人自定位姓名：江萬(wàn)里申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：控制理論與控制工程指導(dǎo)教師：熊蓉褚健20080601浙江大學(xué)碩士學(xué)位論文AbstractR譬Sdfloealizationisalsocited鵲poseestimationproblem在揖aprocessinwhichrobotgaduanygainsitsestimationofselfposebasedOnagiven

2、mapandsensordataLocalizationisacknowledgedasOllethemostfundamentalabilitiesforamobilerobotasabasicftmetiontoimplementotlaertasksHowevertlacnoiseinsensordatathelocalsimilaritiesofsensing，andthodymmiefeaturoofrealenvironme

3、ntbCCOllllethemostchallengesofmobilerobotlocalizationproblemThispaperinvestigatesintotheaetivolocalizationproblemforasinglcrobotinclymmieindoorenvironmentBased011massiveread噸ofresearchpapersandacomparisonofcurrentlocaliz

4、ationandfiltermethods，it8etOnparticlefilterandinfofilteringmethodtosolvetheproblemThemainworkinthispaperis：Firstly，investigatedandimplementedmobilerobotlocalizationmethodbasedOnparticlefilter；experimentedanddiscussedimpo

5、rtantissuesofthealgorithm：particlenumberresamplingmethod，randomparticles，adaptiveparticlefilteringandclymmieenvironmentlocalization；proposedaIllCWmethodtoadjustnoisoparametersinuNatingaprioridistributions，basedOnthedensi

6、tyofparticlecongregationThemethodisprovedtoe1311a110clocalizationspeedandprecisionSecondlyfordyl均mieenvironmentbasedontheideaofoutlinerrejectionmodelingof$e1180rmeasurementandanlll3knownobjectrejectionalgorithma他implemen

7、tedThirdlyproposedasemiactivelocalizationmethod：itrandomlygeneratestargetpositionbasedOnasetcycle，tokeeprobotoutofplaceswherehighlocalsimilaritiesmayexistSincepamr]anningisthebasicsofaetivolocalizationthispaperalsoinvest

8、igatedmobilerobotlocalpathplanningproblemandproposedanewmethodbasedOnviItIlalsideslipforce，whicheffectivelyleadsrobotoutoflocaltrapinclusteredenvironmentAlgorithmsaireimplementedinsimulationwithdatacollectedvirtually蠲wel

9、l鵲fromrealindoorenvironment11把proposedlocalizationalgorithmbasedOnparticlefilterisillustratedtobeeffectiveandrobustinsolvingclassicproblems嬲tracking，globallocalizationandkictnapring，infilteringoutunknownobjectmeasurement