34th Annual International Conference of the IEEE EMBSSan Diego, California USA, 28 August - 1 September, 2012
TrainingStrategiesforaLowerLimbRehabilitationRobotBasedon
ImpedanceControl*
JinHu,ZengguangHou,FengZhang,YixiongChenandPengfengLi
Abstract—Thispaperproposesthreetrainingstrategiesbasedonimpedancecontrol,includingpassivetraining,damping-activetrainingandspring-activetraining,fora3-DOFlowerlimbrehabilitationrobotdesignedforpatientswithparaplegiaorhemiplegia.Controllerswithsimilarstructurearedevelopedforthesetrainingstrategies,consistingofdualclosedloops,theouterimpedancecontrolloopandtheinnerposi-tion/velocitycontrolloop,knownasposition-basedimpedancecontrolmethod.Simulationresultsverifythatposition-basedimpedancecontrolapproachisfeasibletoaccomplishthetrainingstrategies.
I.INTRODUCTION
Plentyofcommondiseasesandaccidentsmayleadtoparaplegiaorhemiplegia,accordingly,thenumberofthesepatientsisgrowinghuge.Comparedtotraditionalphysio-therapy,rehabilitationrobotcanreducethecost.Besides,itcanprovidevarioustrainingstrategies,broadlyclassi edintopassivetrainingandactivetraining,whichbothhavebeenproveneffectivetorehabilitationbyclinictrials[1],[2],[3].Duringrehabilitationtraining,affectedlimbsarerequireddirectcontactwithrobot.Forthebene tofsecurity,forcecontrolforrehabilitationrobothastobeconsidered.Themostextensivelyusedapproachisimpedancecontrol.
Y.Yangetal.proposedaposition-basedimpedancecontrolmethodforanarmrehabilitationrobottoachieveacertaincomplianceandsecurityduringpassivetraining[4].E.G.Caoetal.adoptedanimpedancecontrolapproachforasit-to-standtrainersysteminlowerlimbrehabilitationtoassurethatpatientssafelyandeffectivelystoodupintheirnormalandcomfortableposture[5].Animpedancecontrollawwasalsoemployedin[6]toa3-DOFupperlimbrehabilitationrobotinhuman-robotinteractionsforitsadvantagesintermofsafetytoful lltwooftherehabilitationmodes.
Overall,impedancecontrolusedforrehabilitationrobotmainlyfocusoncompliantbehavior,andthereislittleliteratureinvolvingelaboratetrainingstrategiesonthebasisofimpedancecontrol.Inthisstudy,trainingstrategiesusingposition-basedimpedancecontrolfora3-DOFlowerlimbrehabilitationrobotdesignedforpatientswithparaplegiaorhemiplegiaaremoreelaboratelydescribed.
*ThisworkissupportedinpartbyNationalNaturalScienceFoundationofChina(Grant#61175076)andbySci.&Tech.ProgramforSupervisoroftheRecipientofOutstandingDoctorateDissertationofBeijing(Grant#YB20108000103)
JinHu,ZengguangHou,FengZhang,YixiongChenandPengfengLiarewithStateKeyLaboratoryofManagementandControlforComplexSystems,InstituteofAutomation,ChineseAcademyofSciences,Beijing100190,Chinaustbhujin@163.com,
II.LOWERLIMBREHABILITATIONROBOT
Themechanicalstructureofthe3-DOFexoskeletonrobotforlowerlimbrehabilitationisshowninFig.1.Threejointscorrespondtohipjoint,kneejointandanklejointofalowerlimb.Anabsoluteencoder,arelativeencoderandatorquesensorareinstalledateachjointtorecordthejointangle,angularvelocityandtorque,respectively.Duringrehabilitationtraining,lowerlimbisattachedtotherobot,andtheentiresystem,robottogetherwiththelowerlimbattached,canbeconsideredasathree-linkstructure.Forsimplicity,itisassumedthatthemassofthesystemisevenlydistributed,andthesecondjointbelowistreatedasasinglelink,sincethethirdjointandlinkhavelittlein uenceontheend-pointtrajectory.Therefore,thesimpli edmodelofthesystemisatwo-linkstructureshowninFig.

2.
Fig.1.Mechanicalstructureofrehabilitationrobot
Fig.2.Simpli edstructureofsystem
A.KinematicAnalysis
FromFig.2,forwardkinematicsofthesystemcanbedirectlydeducedas xl1cosq1+l2cos(q1+q2)=(1)yl1sinq1+l2sin(q1+q2)wherel1,l2representthelengthoflink1andthatoflink2respectively,andq1,q2representtheangleofjoint1andthatofjoint2,andx,yrepresenttheend-pointpositioninCartesianspacecoordinates.Inversekinematicscanbederivedfrom(1),undertheconstraintthatq2<0,representedas
22 x2+y2 l l arccosq212
(2)= ylsinq .arcsin22 arctan1q122
x+y
Bydifferentiating(1),Jacobianmatrixcanbewrittenas
l1sinq1 l2sin(q1+q2) l2sin(q1+q2)J=.(3)
l1cosq1+l2cos(q1+q2)l2cos(q1+q2)
hou@compsys.ia.ac.cn,zfeng1203@yahoo.com.cn,yourfriendcyx@163.com,lipengfeng007@163.com
Ifq2=0,det(J)=0,i.e.rank(J)<2,makingthesysteminsingularcon gurations,whichshouldbeavoidedbysetting
Intelligent Exoskeleton for Lower Limb Rehabilitation...power-assist control based on biological feedback....J., “Locomotor training after human spinal cord...
lower limb orthoses for gait rehabilitation_专业...to control knee impedance, is a prime example33...Control strategies and algorithms for portable ...
lower limb gait rehabilitation training method based on impedance control.The ...robot target impedance reflect the system energy consumption,Perhaps has strong...
暂无评价 4页 免费 Electrical impedance in ... 暂无评价 6页 2财富值...Omnidirectional lower limbs rehabilitation training robot is designed based on ...
training in stroke rehabilitation a realistic option...stimulation strategies for upper limb recovery [36...impedance control concept [38,39], a patent ...
Training? Robot? (ART)? based? on? a? cable...rehabilitation? treatment? on? musculoskeletal? ...control?strategies?for?the?passive?force?servo?...
38 Chapter 5 Design of Supplementary Training Software Based on VR... 39 ...54 6.3 Impedance Control Strategy Study of Upper-limb Rehabilitation Robot ...
training has become an established rehabilitation ...kinematics of our lower limb rehabilitation system....Kruger, “Gait rehabilitation machines based on ...
lower limb motor control strategies in elderly ...Programmable mechanical impedance ?EMG based active ...25 Rehabilitation Robot FRIEND IIThe General Concept...
Patient cooperative strategies for robot-aided treadmill training: First ...Artificial Limb Exoskeleton Based on the Control of EMG for Rehabilitation ...
热门文档