Page 1 of 1

SO and JRA output discrepancies (OpenSim v33 vs v41)

Posted: Wed Feb 24, 2021 8:31 am
by giorgiod
Dear all,

While updating some MATLAB codes (from v3.3 to v4.1) to run the classic simulation pipeline (i.e., IK, ID, SO and JRA) and checking for results consistency (between sw versions), we observed some non-negligible differences in the JRA output (~10% in correspondence of the second characteristic peak. See picture below). On the other hand, IK and ID results were almost identical between software versions.

JRA.png
Knee Joint Reaction force
JRA.png (26.32 KiB) Viewed 122 times

Could this be uniquely attributed to the minimum muscle activation that in v4.1 is forced to be no less than 0.01 (while earlier could be 0)? Or else?

Note that all input files (c3d, trc and mot files) were the same. Both model and mot files were converted following the instructions and methods suggested here and/or previous forum posts.

Any help or insight would be greatly appreciated.

Cheers,
Giorgio

PS: Some additional pictures from SO analysis...
vasLat.png
Vastus Lateralis activation
vasLat.png (16.67 KiB) Viewed 122 times
vasLat_force.png
Vastus Lateralis force
vasLat_force.png (18.1 KiB) Viewed 122 times
tfl.png
Tensor Fasciae Latae force
tfl.png (22.43 KiB) Viewed 122 times