simtk.org

Hello everyone,

I have a general question regarding SO. When running SO, we need to add reserve actuators for each coordinate with certain 'optimal_force' value. After running, if I started to decrease the 'optimal_force' value, it seems that, sometimes it still able to solve the problem, but muscles forces become higher while reserve actuators become less active. Also, at the same time, "Performance" value becomes larger.

1) Just wondering, when I can know that simulations got some optimization values? The following link suggests some tips for the CMC. Again the threshold values are task specific.

http://simtk-confluence.stanford.edu:80 ... d+with+CMC

2) The following link also suggests some good tips:

https://simtk.org/plugins/moinmoin/mode ... n/ToolTips

It is mentioned that larger "Performance" means SO is struggling to get solutions. Is there any limit (say,10 or 20) of how large i.e. I need to increase the 'optimal_force' for actuators? Or, again it depends on different task/motion? I found that sometimes SO solve with very high "Performance" values with high muscle forces. Just thinking any cut-off values or for how large "Performance" values I should stop.

Thanks in advance!

Whether a model/simulation parameter (e.g., optimal_force) or output (e.g., peak reserve actuator torque) is appropriate depends on your model, research question, etc. The magnitudes of the reserve actuator torques, for example, might depend on how closely the model represents the subject, what motion is being studied, the quality of the experimental data, etc. Whether the magnitude of a reserve actuator torque is acceptable ultimately depends on what you're trying to learn from your simulations. For example, high reserve torques in the shoulder are unlikely to be important for a study on the ankle. You might be interested in the following paper, which discusses these issues in detail and provides examples you may find helpful:

Hicks, J.L., et al. Is my model good enough? Best practices for verification and validation of musculoskeletal models and simulations of movement. ASME Journal of Biomechanical Engineering 137(2):020905, 2015. http://nmbl.stanford.edu/publications/pdf/Hicks2015.pdf