Open chrisdembia opened 4 years ago
This was a repeated issue during the March 2020 workshop.
Having some feedback/output on what's being done behind the scenes could also help.
For instance, the radians issue was found because of the tracke_states.sto
file. While the input had inDegrees=yes
, the output file had inDegrees=no
even though all the values were the same.
I'm not sure what is meant by the output file. Was that the MocoSolution STO file?
Oops, output file = tracked_states.sto
file. That's a typo.
Oh! That's generated by MocoTrack during preprocessing. I forgot. That's a useful clue. This is a high-priority issue.
I do not understand where the issue is, given that here is where process()
is invoked:
and this is where the angle conversion happens in process()
:
@psbiomech have you used IK solutions as the state tracking reference in Moco?
@carmichaelong it might be helpful to have a set of files that reproduce the problem. It probably warrants it's own GitHub issue; if you have some time to reproduce and create the issue that would be helpful.
Definitely. Will plan on trying to get a minimal example for C++ this week.
Main issues are changing column labels and converting to radians.