基于時(shí)滯多傳感器數(shù)據(jù)的信息融合濾波.pdf

1、分類號(hào)UDC密級(jí)公開Z秀廬夕擎碩士研究生學(xué)位論文基于時(shí)滯多傳感器數(shù)據(jù)的信息融合濾波申請(qǐng)人：學(xué)號(hào)：培養(yǎng)單位：專業(yè)：研究方向：指導(dǎo)教師：完成日期：王光輝2100776電子工程學(xué)院控制理論與控制工程多傳感器信息融合孫書利教授2013年4月15日黑龍江大學(xué)碩士學(xué)位論文AbstractWiththerapiddevelopmentofnetworkcommunicationtechnologyandcomputertechnologyavastn

2、umberofmultivsensorsystemsemceeforvariousapplicationsNetworktransmissiontechnologiesarewidelyusedinaerospace，militaryindustrialandotherfieldsInpracticalapplicationsystems，howeverdelayphenomenonshouldnotbeignoredduetothev

3、ariabilityoftransmissionchannelsandunreliabilityofcommunicationFordelayedsystems，traditionalKalmanestimationtheoryisnotapplicableByusingtheprojectiontheoryandweightedfusionalgorithminthelinearminimumvariancesense，thefoll

4、owinginnovativeresearchresultsareproposedforthefusionestimationproblemsofdelayedsystems，F(xiàn)ormultisensorlineardiscretesystemswithconstanttimedelaytheconstantdelaylineardiscretesystemisconvertedtoanewsystemwithoutdelaysthro

5、ughthenonaugmentedmethodFirstlyoptimalKalmanestimatorsarepresentedforsinglesensorsystemThelocalestimationerrorcrosscovariancematricesofanytwosubsystemsarcderivedFurtherdistributedweightedinformationfusionestimatorisgiven

6、Formultisensorlineardiscretesystemswithstochasticoutofsequencetracks，therandomdelaysOccurinthetransmissionprocessfromlocalstatefiltertothefusionprocessingcenteLThelatestlocalestimateswhichachievethefusioncenterattheprese

7、ntarefusedIfthereisnolocalestimatereceivedinthepresentmoment，thelocalfilteratlastmomentwillbeappliedbypredictionFurthermore，theCIalgorithmwhichavoidscalculationofcrosscovariancematrixisusedforfusionestimationFormultisens

8、orlineardiscretesystemswithrandomoutofsequencemeasurement，theme孫urememdelaysinsystemmodelisdescribedbyaBernoullidistributedrandomvariableFkstly，theoptimallinearfilterinlinearminimumvariancesenseispresentedforsinglesensor