Page 1 of 1

RRA crashing problems

Posted: Wed May 28, 2008 8:57 am
by mdemers
Hi all,

I'm trying to complete RRA and CMC on a jumping trial and keep getting the error below at the beginning or RRA.

"ERROR - Evaluating GCVSpline at 0.008327 outside its domain [-0.185833, 0]"

This occurs just after "Padded kineticsStore to start time of 0.064167"

Per Kat's recommendation I edited the grf.mot file to have the same start and end times as the IK solution, but this didn't work. Has anyone encountered this before and have any insight?

thanks,
Matt

RE: RRA crashing problems

Posted: Wed May 28, 2008 9:04 am
by mql
Hi Matt,

What happens if you try to start RRA at a different time (not the very first time in the input files)? RRA/CMC have to prepend data to the beginning of your data in order to do smooth spline fits. That's probably the "padded kineticsStore". It does this by mirroring your own data. Sometimes this is not a smooth fit and causes problems.

So, do you still get the error if you start RRA at bit later (try even just 50 ms) in the data? Some time when perhaps there's unlikely to be an inflection point when it fits the data?

- May

RE: RRA crashing problems

Posted: Wed May 28, 2008 9:34 am
by mdemers
Thanks May,

I tried changing my start time from 0.85 to 0.9 and 1 and neither worked. 0.9 and 1 are definitely after foot strike. I keep getting

Padded kineticsStore to start time of 0.341667
Padded kineticsStore to start time of 0.091667

Exception:
ERROR - Evaluating GCVSpline at 0.008327 outside its domain [-0.158333, 0]

0.341667 is the initial time in my IK and grf.mot files, but 0.091667 is definitely outside the time range of the data I'm sending to RRA. Could this be associated with the problem?

- Matt