预览加载中，请您耐心等待几秒...

基于ROC曲线优化的车辆行驶状态估计BP模型.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于ROC曲线优化的车辆行驶状态估计BP模型 Abstract Inthispaper,weproposeanoptimizedBP(Backpropagation)modelforvehicledrivingstateestimationbasedonROC(ReceiverOperatingCharacteristic)curveanalysis.TheobjectiveofthisresearchistoimprovetheaccuracyofvehicledrivingstateestimationbyoptimizingtheparametersintheBPmodelusingtheROCcurve.TheresultsofourstudyindicatethattheproposedoptimizedBPmodelprovidesbetteraccuracythanthestandardBPmodelinpredictingvehicledrivingstates. Introduction Vehicledrivingstateestimationisanimportanttaskinthefieldofintelligenttransportationsystems(ITS).Itinvolvesmonitoringthebehaviorofavehicleandpredictingitsnextstate.Thisinformationcanbeusedtoenhancesafetyandefficiencyofvehiculartraffic.Oneofthetechniquesforvehicledrivingstateestimationismachinelearning,whichinvolvesdevelopinganestimationmodelusinghistoricaldata.BPneuralnetworkisonesuchpopularmachinelearningtechnique. However,developinganaccurateBPmodelrequiresselectingtheoptimalsetofmodelparameters.Thisinvolvesatrade-offbetweenminimizingthetrainingerrorandreducingthegeneralizationerror.TheROCcurveanalysisprovidesatechniqueforselectingtheoptimalsetofmodelparameters.Inthisstudy,weproposeanoptimizedBPmodelforvehicledrivingstateestimationbyanalyzingtheROCcurve. Methods Thedatausedinthisstudyisobtainedfromapubliclyavailabledataset,theNaturalisticDrivingStudydataset.Thedatasetcontainsdatacollectedfrominstrumentedvehiclesdrivenbyvolunteersinthenaturalenvironment.Thedataconsistsofvariousfeaturesrelatedtovehicledrivingstatessuchasvehiclespeed,acceleration,steeringangle,etc.Thedatasetissplitintotrainingandtestingdatasets,with70%ofthedatabeingusedfortrainingandtheremaining30%fortesting. TheBPmodelisimplementedusingPython’sscikit-learnlibrary.Themodelconsistsofaninputlayer,oneormorehiddenlayers,andanoutputlayer.Thenumberofnodesintheinputlayerisequaltothenumberoffeaturesinthedataset.Thenumberofnodesintheoutputlayerissettotwo,representingthetwodrivingstates:normalandaggressive.Thenumberofhiddenlayersandnodesineachlayerisvariedtoop