simtk.org

Hi,
I want to modelize a squat movement using OpenSense so with measurements acquired with my experimental IMUs. I choose the Rajagopal 2015 model but it seems that the IMU sensor located on the pelvis (IMU base) does not move, the legs instead seems to going up (like it is doing a fake jump). I add also same photo to showing to you what i'm talking about.
The angle reached by the hip and knee flexion seems corrects but i'm not understanding why the pelvis imu doesn't move.
I've already try to change the model and using some suggested for the squat measures but it does always the same. Is it some specific setting that i need to change?
Thank you for the answer

Hello,

I think the issue here is that you're expecting the IMU inverse kinematics to track translations, however IMUs typically do not provide position/translation information, only rotations/orientations. As such a squat and a fake jump may look similar. Depending on your research/question that may or may not be a problem, if it is then you'll have to use other mechanisms to obtain/enforce translations.

Hope this helps,
-Ayman

I'm sorry i probabily don't get it.
If the IMU provide just the orientation how the model is able to understand that the others sensors needs to move/rotate with respect to the pelvis one and not with respect to another one?
I've already tried to change what is called the base IMU choosing another one but it seems that nothing changes.
Opening the file .mot generating by the IK it is clear that the pelvis translations remains the same (default) value during all the motion. How can i change the setting for having the foots in the defaults position, so they will remain still and say to the model to unlock the pelvis traslations?

Thank you for your help,
Cristina