Bayesian Filtering From Kalman Filters to Particle Filter and Beyonds

《Bayesian Filtering From Kalman Filters to Particle Filter and Beyonds》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、MANUSCRIPT1BayesianFiltering:FromKalmanFilterstoParticleFilters,andBeyondZHECHENAbstractInthisself-containedsurvey/reviewpaper,wesystem-IVBayesianOptimalFiltering9aticallyinvestigatetherootsofBayesianfilteringaswellasitsrichIV-AOptimalFiltering....................

2、.10leavesintheliterature.StochasticfilteringtheoryisbrieflyreviewedIV-BKalmanFiltering.....................11withemphasisonnonlinearandnon-Gaussianfiltering.FollowingIV-COptimumNonlinearFiltering..............13theBayesianstatistics,differentBayesianfilteringtechnique

3、sarede-IV-C.1Finite-dimensionalFilters............13velopedgivendifferentscenarios.UnderlinearquadraticGaussiancircumstance,thecelebratedKalmanfiltercanbederivedwithintheVNumericalApproximationMethods14Bayesianframework.Optimal/suboptimalnonlinearfilteringtech-nique

4、sareextensivelyinvestigated.Inparticular,wefocusourat-V-AGaussian/LaplaceApproximation............14tentionontheBayesianfilteringapproachbasedonsequentialMonteV-BIterativeQuadrature...................14Carlosampling,theso-calledparticlefilters.ManyvariantsoftheV-CM

5、ulitgridMethodandPoint-MassApproximation..14particlefilteraswellastheirfeatures(strengthsandweaknesses)areV-DMomentApproximation.................15discussed.RelatedtheoreticalandpracticalissuesareaddressedinV-EGaussianSumApproximation..............16detail.Inaddit

6、ion,someother(new)directionsonBayesianfilteringV-FDeterministicSamplingApproximation.........16arealsoexplored.V-GMonteCarloSamplingApproximation.........17IndexTermsStochasticfiltering,Bayesianfiltering,V-G.1ImportanceSampling..............18Bayesianinference,parti

7、clefilter,sequentialMonteCarlo,V-G.2RejectionSampling................19sequentialstateestimation,MonteCarlomethods.V-G.3SequentialImportanceSampling........19V-G.4Sampling-ImportanceResampling.......20V-G.5StratifiedSampling................21“Theprobabilityofanyeve

8、ntistheratiobetweentheV-G.6MarkovChainMonteCarlo...........22valueatwhichanexpectationdependingonthehappeningoftheeventoughttobecomputed,andthevalueoftheV-G.7H