Do NaN values affect Inverse Kinematics Solution?

Provide easy-to-use, extensible software for modeling, simulating, controlling, and analyzing the neuromusculoskeletal system.
POST REPLY
User avatar
Shaun Fickling
Posts: 13
Joined: Fri Feb 24, 2012 11:48 am

Do NaN values affect Inverse Kinematics Solution?

Post by Shaun Fickling » Wed Dec 04, 2013 8:07 am

Hi

For some of my markers I have large sections of position data missing. Is it alright to set the marker positions at these points as NaN or will that add large errors to the other markers on the model during Inverse Kinematics?

Would it be better instead to remove such markers from the simulation altogether or just observe the trial over a time-window when data is available for all of the markers?

Regards,
Shaun Fickling
Top
User avatar
David John Saxby
Posts: 83
Joined: Mon May 09, 2011 8:39 pm

Re: Do NaN values affect Inverse Kinematics Solution?

Post by David John Saxby » Thu Dec 05, 2013 7:03 pm

Hi Shaun,

I don't think this will "add" error, but a NAN for a marker will be dropped from the evaluation of IK's objective function for the frames were a NAN is present. The result will be a global solution for model general coordinates which does not include that marker(s). If that marker(s) was important to establishing mobilities then you may have a problem, if not...not :shock:

DJS
Top
POST REPLY