Using rolling constraint rather than ground contact model for CMC?

[AUDE LEFRANC]

Hello All!

Is it possible to use a rolling constraint during CMC rather than a ground contact model? If so, what is the best way to add that to the model? I've been using a HuntCrossley based GCM and it seems to be producing pretty good vertical GRFs, but the ML and AP GRFs still seem pretty far off regardless of what friction and stiffness parameters I use. I'm open to any suggestions y'all might have to produce better GRFs from CMC.

Thank you in advance!

[John Davis]

This kind of approach was used in Hamner et al.'s simulations of muscle contribution to running. The SimTK respository here has the models and CMC setup files they used--those might be a good starting point! A different solution might be to ditch CMC and use Moco instead to explicitly track the GRFs alongside the kinematics.