simtk.org

Hey guys,

referring to Steele at al. [Steele, Katherine M., et al. "Compressive tibiofemoral force during crouch gait." Gait & posture 35.4 (2012): 556-560.] it is possible to include different weights c_i for the muscles i in static optimization. Thus, the objective function becomes min{sum(c_i*(a_i)^2)} where a_i is the activation level of each muscle.
I couldn't find any possibility to implement such weights in OpenSim - neither in the GUI nor in the XML-file. In the simtk-confluence-User's Guide the objective funtion is given without any weights as well.
Since I wanna find tibiofemoral contact forces just like Steele et al. and their methodology led to good, reasonable results, it would be great to use the same methods, i.e. the same weights for static optimizaiton.

So, my question would be if and how I can include muscle weighting constants in the static optimizaiton tool.

Thanks for your time and consideration.
Best,
Felix

Hi Felix,

We modified the OpenSim source code in order to weight the activations for Static Optimization. This was just about the time when the transition from OpenSim 1.9.1 to 2.0 occurred which involved a lot of changes to the underlying code and the weights were never integrated with the new source.

However, I have recently developed a new OpenSim tool that will allow you to weight muscles or couple together the activation of multiple muscles for Static Optimization. It is currently functional in the source code and we are planning to integrate it with a release of OpenSim next fall.

Regards,
Kat Steele

Hi Katherine,

This tool that you have developed sounds very interesting! You mentioned that it is available in the source code. If I were to download the source code, would I be able to add it to OpenSim as a plugin or through another method?

What I would really like is to be able to use this tool while performing CMC. Would you be able to give us the name of it or where to find it in the source code?

Any advice would be a huge help! Looking forward to using your tool.

Thanks,

Alex