simtk.org

Hi All

I am currently using the Static Opt to estimate the muscle forces for stair climbing. For normal subjects, the estimated muscle forces seems reasonable. However, for injured subjects, the muscles forces are extremely small due to the large force in the appended static optimization actuators. I am wondering if there is a mechanism to decrease the required extra actuator forces.

Hi Kang,

You can decrease the strength of the residual actuators be decreasing their optimal force value. You can find that tag in the xml file for the model. You should run inverse dynamics first and look at the required generalized forces for the 6 pelvis degrees of freedom. The maxima/minima of these force trajectories will tell you how strong you have to make your residual actuators.

High residuals are a sign of mismatch between applied external force, their point of application, and the measure/fit model kinematics. Exploring residuals should start with making sure that you are actually applying the correct external forces, to the correct point, in the correct orientation, and to the correct body.

You should also consider running Reduce Residuals Algorithm (within the CMC tool) before doing static optimization calculations. This will hopefully yield a model that is more consistent with the measured forces and kinematics, thereby reducing the pelvis residuals in static optimization.

Thanks Matt for the suggestions.
I will try. On the other hand, there is another way to improve the performance is to improve the muscle path to get better moment arm. By doing that, the result could be more favorable.