August 17, 2009
Announcements
Announcements
- Move the meeting to Friday at 2:15p?
- Some new faces: Joey, Shalv, and Sharon.
- Seminars: Phys - wed@12p; Emory Spinal Chord - wed@8:30a; Robotics - ???
- Stroke papers: Reisman and Scholz 2003, Dewald 2001, Dewald 1995
Presentation
Presentation
by Keith van Antwerp
Todorov, Jordan. 2002. Optimal Control as a Theory for Motor Coordination. Nature Neuroscience.
- Lots of degrees of freedom in the biological system
- Biological systems are capable of reliably achieving high-level goals
- Variability occurs on the uncontrolled manifold
- Enforcing end-point trajectories often removes the variability observed in biology
- Stochastic Optimal Control explains this better
- Assumptions
- Actuator noise increases with magnitude of signal
- An internal forward model exists
- The sensors have noise and the feedback is delayed
- The system has a method of penalizing effort
- Assumptions
- The uncertainty of the system is projected onto the redundant degrees-of-freedom
- Coupling of the control signals helps counter-act addition of "noise"
Paper discussion
Paper discussion
- Do muscle synergies produce task-level functions, or are they an artifact of a sophisticated analysis?
- Are muscle synergies innate or learned?
- How many muscle synergies are required for task performance?
- Theory of sparse coding and how it applies to neuromechanics.
- Variability of nervous and mechanics.
- Optimal control
- Pathologies
- Sensory information.