Transforming External Loads w/ Changing Kineamtics

[Brecca Gaffney]

Hi Folks,

I am working on assessing how perturbations in hip kinematics alter joint loading during dynamic activities. Currently, I am generating a perturbed kinematics file by altering the hip abduction angle from RRA. To maintain dynamic consistency in CMC/Analysis tool, I am expressing the GRF in the local body of the foot frame by specifying the perturbed kinematics file within the external loads. However, I am unsure of the effect that this is having on my results, as there is no difference in JRF between the baseline RRA vs. perturbed kinematics specified in the kinematics file (the desired kinematics stays the same in the main CMC/analysis tools is the perturbed kinematics).

Two questions:

1) When specifying the kinematics file in the external loads to express the force in the local coordinate system of a specified body (e.g. calcn_r), does this account for translation of the body or is just a rotational transformation?

2) If not, is there functionality in the API to account for the translation between perturbed kinematics and measured GRF (see figure)? It seems like this would be accounted for in the transformPointExpressedInGroundToAppliedBody() command but I am unsure why there would be no difference in 'transformed' external loads (i.e. using the perturbed kinematics) and the baseline external loads.

Picture1.png (321.6 KiB) Viewed 207 times

Thanks for your your help.

[jimmy d]

Hi, Brecca-

Ajay's post from a few years back on this may prove useful (if you hadn't seen it already).

My understanding is that, when you specify Kinematics for External Loads, the point-of-force application on the body will be maintained even as RRA alters the model kinematics. If you don't specify Kinematics for External Loads, the point will stay the same (in the global).

To maintain dynamic consistency...

I think it is going to be hard to do that since you are making a trial that will be intentionally inconsistent.

For the study that you are describing, you could possibly generate a foot contact model (or add foot contact objects), with parameters found from your known/collected trials, then systematically perturb a forward simulation with varying kinematics. It would be something like in the this tutorial and paper

-J