Adaptive, Fast Walking in a Biped Robot under Neuronal Control and Learning 英文参考文献.docVIP

Adaptive,FastWalkinginaBipedRobot underNeuronalControlandLearning Poramate Manoonpong1[,Tao Geng2[,Tomas Kulvicius1,Bernd Porr3,Florentin Wo¨rgo¨tter1,2* 1 Bernstein Center for Computational Neuroscience, University of Go¨ttingen, Go¨ttingen, Germany, 2 Department of Psychology, University of Stirling, Stirling, Scotland, UnitedKingdom,3DepartmentofElectronicsandElectricalEngineering,UniversityofGlasgow,Glasgow,Scotland,UnitedKingdom Humanwalkingisadynamic,partlyself-stabilizingprocessrelyingontheinteractionofthebiomechanicaldesignwith its neuronal control. The coordination of this process is a very difficult problem, and it has been suggested that it involvesahierarchyoflevels,wherethelowerones,e.g.,interactionsbetweenmusclesandthespinalcord,arelargely autonomous, and where higher level control (e.g., cortical) arises only pointwise, as needed. This requires an architecture of several nested, sensori–motor loops where the walking process provides feedback signals to the walker’ssensorysystems,whichcanbeusedtocoordinateitsmovements.Tocomplicatethesituation,atamaximal walking speed of more than four leg-lengths per second, the cycle period available to coordinate all these loops is rathershort.Inthisstudywepresentaplanarbipedrobot,whichusesthedesignprincipleofnestedloopstocombine theself-stabilizingpropertiesofitsbiomechanicaldesignwithseverallevelsofneuronalcontrol.Specifically,weshow howtoadaptcontrolbyincludingonlinelearningmechanismsbasedonsimulatedsynapticplasticity.Thisrobotcan walk with a high speed (.3.0 leg length/s), self-adapting to minor disturbances, and reacting in a robust way to abruptlyinducedgaitchanges.Atthesametime,itcanlearnwalkingondifferentterrains,requiringonlyfewlearning experiences.Thisstudyshowsthatthetightcouplingofphysicalwithneuronalcontrol,guidedbysensoryfeedback fromthewalkingpatternitself,combinedwithsynapticlearningmaybeawayforwardtobetterunderstandandsolve coordinationproblemsinothercomplexmotortasks. Citation:ManoonpongP,GengT,K