Forward simulation without CMC and GRF's

[Anthony Ligouri]

Is there a way to use the kinematics of the torso to drive a forward simulation of arm swing? Here is why I ask...

My lab is interested in studying the differences in arm swing kinematics between normal human gait and arm swing that is completely passive, i.e. arm motion that is only the result of movement of the upper body without any muscle action, similar to a driven pendulum.

My initial thoughs on how to make this comparison were as follows: From our gait lab mocap data I could use OpenSim to make a scaled model of our subject and calculate the inverse kinematics. This, obviously, would be the armswing kinematics for normal gait. From inverse kinematics, I could also get the translations and rotations of the torso. To recreate passive arm swing, I was hoping to use the torso kinematics in a forward simulation with a model that just contained torso, upper arm, and lower arm bodies and associated musculature where I could simply drive the motion of the torso and see how the mass properties of the arm reacted and record the kinematics.

Unfortunately, it seems that the forward dynamics tool is set up to only take controls in the form of muscle excitations, which I do not have. Nor do I have GRF's for my gait data, so I am not able to do inverse dynamics, CMC, etc... Any ideas? Thanks all.