Hello,
I am trying to use Induced Acceleration Analysis to calculate muscle potentials for a subject rising from a chair. However, the resulting potentials in the output file (model_InducedAccelerations_center_of_mass.sto) are incredibly large in all 3 directions. Previous subjects' data is orders of magnitude smaller than the data for this subject. Does anyone have any ideas or suggestions on how to troubleshoot the problem or what may be the cause of it?
Thanks,
Rachel
Unrealistic Induced Acceleration Potentials
-
Rachel Baker
- Posts: 5
- Joined: Wed Aug 20, 2014 2:22 pm
-
Matt DeMers
- Posts: 16
- Joined: Mon Oct 08, 2007 8:36 pm
Re: Unrealistic Induced Acceleration Potentials
Hi Rachel,
The usual culprit is a mismatch in data sources. For IAA, you must provide kinematics, ground reaction forces, and actuations as separate files. You have to be sure that the model, kinematics, grf, and actuation files are from the same trial and synchronized. If they aren't, you're in effect creating a dynamically inconsistent system and the results will be all over the place. Most of the time, mismatch between these files is the culprit for people I've helped. For example, mistakenly using trialA kinematics with trialB grf or trialC actuation trajectories.
Assuming you've got all the ducks in a row and it's not a simple mismatch, what is deferent between this subject and your "previous subjects"? Are you using the same version of opensim? Same preprocessing/filtering settings?
Matt
The usual culprit is a mismatch in data sources. For IAA, you must provide kinematics, ground reaction forces, and actuations as separate files. You have to be sure that the model, kinematics, grf, and actuation files are from the same trial and synchronized. If they aren't, you're in effect creating a dynamically inconsistent system and the results will be all over the place. Most of the time, mismatch between these files is the culprit for people I've helped. For example, mistakenly using trialA kinematics with trialB grf or trialC actuation trajectories.
Assuming you've got all the ducks in a row and it's not a simple mismatch, what is deferent between this subject and your "previous subjects"? Are you using the same version of opensim? Same preprocessing/filtering settings?
Matt