Torsional knee spring using gait2392 or gait2354

[Hossein Mokhtarzadeh]

Hi all,

I am using this example below to perform a muscle-driven simulation with a torsional knee spring using gait2392 or gait2354 core models:

"Simulation-Based Design to Reduce Metabolic Cost" page (https://simtk-confluence.stanford.edu:8 ... iveDevices).

When I add torsional knee spring (+ Metabolic Probes ) to the gait10dof18musc.osim model, and follow the steps in the example, all is good and I can run CMC, etc. The results are sensitive to spring's parameters.

However, when I use OpenSim examples for gait2392 or gait2354 core models and add torsional knee spring and then do CMC (using all xml files from the the same examples e.g. gait2392 or gait2354 provided), I don't see any differences in muscle activation patterns or metabolic costs. In fact, I change upper_stiffness from 1 to 10.0 or even 20.

It seems I am missing something here. I wonder if anyone has done this example (adding any torsional spring to the core models of gait2392 or gait2354) and performed CMC and whether they have seen changes in activation patterns.

Thank you!
Cheers,
Hossein

[Hossein Mokhtarzadeh]

I think I got why there was no differences. I used the same stiffness and inputs from the example on ankle for the knee spring.

I am just wondering now if RRA and ID can calculate (or consider) other actuators e.g. added torsional knee spring. I could not verify this yet.

Finally, after I perform CMC, how can I verify or calculate CoordinateLimitForce in Matlab via OpenSim API?

Thank you.
Hossein