Results From Calibrating Model
• In general, motion similar to Sprawlita
• Trajectory of COM looks about the same
• Model runs slightly faster than robot (.65m/s vs. .60m/s)
• Whole body and leg GRFs similar in shape, magnitude
•  Model’s legs appear to thrust too long
• For both model and robot locomotion is sensitive to
parameter changes in similar ways:
• Hip stiffness – increases ‘stability,’ slows down
• Stride period – velocity changes with same profile
• Front Leg Angle –  trade off between speed and stability