UT PSY 394U - Unifying abstraction for motor control

Unformatted text preview:

Equilibrium Points: A unifyingabstraction for motor controlXue GuDana H. BallardUniversity of TexasComputing with a physical devicePassive energy storageMechanical complianceDirectly related ideasFaloutsos et al Computers and Graphics25(6) 2001 933-953Gunther and RuderBiological Cybernetics89, 89-106 2003Torres & Zipser J. Neurophys 88 2355-2367 2002The ModelMotor Simulation: Method• Use GD on distance error function for kinematics solution.• Energy consumption optimization with constraints to end effector.Motor Simulation in Experiments• Four fastracks areattached to the arm tocapture movement data• Weights are attached todifferent segments of thearm to test how the endpostures changes.weights attachedfastracksSimulation comparison with human data weights on bicep weights on forearm weights on hand.Movement Execution: Theory• Dynamic skeleton system:• Springs apply torques to the system• is spring stiffness and is viscosity. Withdifferent and , different trajectories aregenerated.KBKBMovement Execution:Implementation• Movement execution• Use springs in Vortex 2.3, a physics simulationpackage, to simulate muscles– Spring controls extension and flexionPlanar Arm ReachingMovement Data Comparison (1)• End effector trajectories fitting human dataooo predictionSpeed profile along xSpeed profile along yxyttxvyvMovement Data Comparison (2)Joint profile comparisonShoulder Elbow WristModel Extension of Walking• Bi-pedal walking:– 4 Via points are defined for one cycle of walking– Every leg has 3 springs, which control bothextension and flexionWalkingWalking Profile ComparisonHuman data from Boston dynamics Model simulation dataHip KneeAnkleWalking IIComplex Movements Generation• High DOF system as complicated as humanbody is hardly tractable• Human body parts move in relativelyindependent manners.• Divide and conquer– Motor routines defined for body parts– Motor synergies are used to combine complexmovementsMotor Synergies• Two or more tasks can be planned at the sametime during motor simulation.– Sitting down: hip reaches to destination– Balancing: center of mass remains above the feet.Hand Grasping Routine• 14 degrees of freedom.• Grasping is done bysetting the EPs insidethe objects.• No high accuracy isrequired duringplanning.Complex MovementsComplex Movements IIGrazianoSegment 1 Segment 2Segment 2Segment 1Plan Segement:SMA, PremotorctxExecuteSegement:Spinal cordProposed OrganizationMovement Control Summary1. Complex Movements Executed in Segments2. Each Segment: Two phases• Offline motor simulation phase! End posture for current task• Online motion Execution phase! Equilibrium point theory! Abstract spring-muscle control model3. Motor synergies• Different routines defined for body parts• Combination for whole body


View Full Document

UT PSY 394U - Unifying abstraction for motor control

Documents in this Course
Roadmap

Roadmap

6 pages

Load more
Download Unifying abstraction for motor control
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Unifying abstraction for motor control and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Unifying abstraction for motor control 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?