Unscented filtering and nonlinear estimation

Unscented filtering and nonlinear estimation

ID:39772832

大小:651.34 KB

页数:22页

时间:2019-07-11

Unscented filtering and nonlinear estimation_第1页
Unscented filtering and nonlinear estimation_第2页
Unscented filtering and nonlinear estimation_第3页
Unscented filtering and nonlinear estimation_第4页
Unscented filtering and nonlinear estimation_第5页
资源描述:

《Unscented filtering and nonlinear estimation》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、UnscentedFilteringandNonlinearEstimationSIMONJ.JULIER,MEMBER,IEEE,ANDJEFFREYK.UHLMANN,MEMBER,IEEEInvitedPaperTheextendedKalmanfilter(EKF)isprobablythemostwidelymostareeithercomputationallyunmanageableorrequireusedestimationalgorithmfornonlinearsystems.However,morespecialassumptionsabouttheformofthep

2、rocessandthan35yearsofexperienceintheestimationcommunityhasshownobservationmodelsthatcannotbesatisfiedinpractice.Forthatisdifficulttoimplement,difficulttotune,andonlyreliableforsystemsthatarealmostlinearonthetimescaleoftheupdates.Manytheseandotherreasons,theKFremainsthemostwidelyofthesedifficultiesa

3、risefromitsuseoflinearization.Toovercomeusedestimationalgorithm.thislimitation,theunscentedtransformation(UT)wasdevelopedasTheKFonlyutilizesthefirsttwomomentsofthestateamethodtopropagatemeanandcovarianceinformationthrough(meanandcovariance)initsupdaterule.Althoughthisisanonlineartransformations.Itis

4、moreaccurate,easiertoimplement,relativelysimplestaterepresentation,itoffersanumberofandusesthesameorderofcalculationsaslinearization.Thispaperreviewsthemotivation,development,use,andimplicationsoftheimportantpracticalbenefits.UT.1)ThemeanandcovarianceofanunknowndistributionKeywords—Estimation,Kalman

5、filtering,nonlinearsystems,requiresthemaintenanceofonlyasmallandconstanttargettracking.amountofinformation,butthatinformationissuffi-cienttosupportmostkindsofoperationalactivities(e.g.,definingavalidationgateforasearchregionforI.INTRODUCTIONatarget).Thus,itisasuccessfulcompromisebetweenThispapercons

6、iderstheproblemofapplyingtheKalmancomputationalcomplexityandrepresentationalflexi-filter(KF)tononlinearsystems.Estimationinnonlinearbility.Bycontrast,thecompletecharacterizationofansystemsisextremelyimportantbecausealmostallpracticalevolvingerrordistributionrequiresthemaintenanceofsystems—fromtarget

7、tracking[1]tovehiclenavigation,anunboundednumberofparameters.Evenifitwerefromchemicalprocessplantcontrol[2]todialysisma-possibletomaintaincompletepdfinformation,thatin-chines—involvenonlinearitiesofon

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。