Hey guys,
I am using induced acceleration to try and compute muscle contributions to segment powers. Since there is no direct functionality to do this, I need to obtain the muscle contributions to the angular accelerations of each segment and their translational accelerations (at the centre of mass of each segment) in order to compute powers.
Using Tim Dorn's induced acceleration plug-in, I only get the contributions to joint accelerations. Does anyone know how I might go about using this data to get segment rotational and translational accelerations?
Cheers,
Nirav
Knee acceleration to tibial acceleration
- Nirav Maniar
- Posts: 32
- Joined: Tue May 12, 2015 11:03 pm
- Thomas Uchida
- Posts: 1789
- Joined: Wed May 16, 2012 11:40 am
Re: Knee acceleration to tibial acceleration
Can you use the BodyKinematics analysis? http://simtk-confluence.stanford.edu:80 ... h+Analyses
- Nirav Maniar
- Posts: 32
- Joined: Tue May 12, 2015 11:03 pm
Re: POSSIBLE BUG Knee acceleration to tibial acceleration
Hey Tom,tkuchida wrote:Can you use the BodyKinematics analysis? http://simtk-confluence.stanford.edu:80 ... h+Analyses
I think I may have found a bug in OpenSim's "kinematic" analysis.
Using Tim's plug in, I calculated the net "EXPERIMENTAL" induced acceleration so that I could cross reference with the experimental data. I double integrated this, obtaining initial positions and velocity from OpenSim "Kinematics" analysis. I then verified my integrated "induced position/velocity" to ensure they matched the experimental values obtained from "kinematics" and they did so perfectly (I did this to generate inputs for body kinematics to get the data I initially posted about).
However, I noticed that the "EXPERIMENTAL" acceleration from Tim's plug in (lets say knee_angle_r for instance) did not match the values provided in the "_dudt.sto" file from OpenSim's "kinematics" analysis. To test which was wrong, I differentiated the joint velocity and position data from "kinematics" and reproduced the "EXPERIMENTAL" value from Tim's plug in, confirming that the issue is within the "_dudt.sto" file. Note that this is not a radians-degrees confusion issue, as the temporal patterns are completely different.
Is this a known bug? And can you confirm you can reproduce the issue with your own data? (issue noted in OpenSim 3.3).
Cheers,
Nirav
- Thomas Uchida
- Posts: 1789
- Joined: Wed May 16, 2012 11:40 am
Re: Knee acceleration to tibial acceleration
Would you please provide a link to Tim's plug-in?
- Nirav Maniar
- Posts: 32
- Joined: Tue May 12, 2015 11:03 pm
Re: Knee acceleration to tibial acceleration
Hi Tom,tkuchida wrote:Would you please provide a link to Tim's plug-in?
You can find Tim's plug-in (IndAccPI) here: https://simtk.org/frs/?group_id=433. Note that it only works on OpenSim 3.2 (32-bit), but I did the kinematic analysis on OpenSim 3.3 (64-bit) to confirm that the issue still exists in the most current version.
Cheers,
Nirav