基于表面肌电信号的下肢康复机器人控制方法分析-analysis of control method of lower limb rehabilitation robot based on surface electromyography.docxVIP

  • 12
  • 0
  • 约5.29万字
  • 约 73页
  • 2018-05-18 发布于上海
  • 举报

基于表面肌电信号的下肢康复机器人控制方法分析-analysis of control method of lower limb rehabilitation robot based on surface electromyography.docx

基于表面肌电信号的下肢康复机器人控制方法分析-analysis of control method of lower limb rehabilitation robot based on surface electromyography

AbstractSurfaceelectromyographysignal(sEMG)isakindofbio-electricitysignal collectedfromthesurfaceofhumanskinbysurfaceEMGelectrodes,whichcan reflecttheprocessofhumanneuromuscularactivity.Notonlyattheareaofclinical medicine,butalsointhefieldofrehabilitationmedicine,doesithaveimportant appliedvalues.Inthispaper,acontrolstrategyforthemovementofkneejointof lowerlimbrehabilitationrobotusedforthepatientswithunilaterallowerlimbmotor impairmentsbecauseofhemiplegicisdesignedbasedonsEMG,tomakethe independentrehabilitationbypatientsthemselvespossible.Meanwhile,thispaper focusesonthekeytechnologiesinthreepartswhicharethefeatureextractionof sEMG,theformationofthemodelofsEMG-themotionofkneejointandthe electromyographiccontrol oflower limb rehabilitation robot.Firstofall,thesEMGfromRF,VMandVLiscollectedrespectivelybythe meansofacquisitionmethodcombiningpostureandmotionbasedontheformation mechanism,thecharacteristics,theinfluencesandthecompositionofnoiseofsEMG. Meanwhile,theangleofkneejointismeasurementinadirectway.Afterwards,RMS, iEMG andMPFthreefeatures valueareextracted to analysis themotion of kneejoint. Attheendofthisaspect,RMSofsEMGisusedtoanalysisthecharacteristicsofthe sEMG fromquadriceps.BasedonthecorrelationbetweensEMGandthemotionofkneejoint,themodel ofsEMG-locationisestablished,byusingtheeigenvaluesfromsEMGastheinput ofBPneuralnetwork,andtheanglesignalfromkneejointastheoutputofneural networks.ThemodelofsEMG-velocity(acceleration)isalsoestablished,byusing sEMG eigenvalues andits first (first andsecond) order derivative together as the input ofneuralnetwork,andthefinalvelocity(acceleration)ofthekneejointastheoutput ofneuralnetwork,basedonthemodelofsEMG-location.Thequantitative identificationtothelocation/velocity/accelerationofkneejointusingtheeigenvalues ofsEMG is achieved.Finally,apassivecontrolstrategyoflowerlimbrehabilitationrobotisdesigned intwowaysofthresholdcontrolandpatternrecognitiontakeadvantageofthe platformoflowerextremityrobot,toachievethecontroltothemovementof

您可能关注的文档

文档评论(0)

1亿VIP精品文档

相关文档