simtk.org

Sorry, been busy/preoccupied.
You set things up how I would have (at least initially), but I have think that how the board is part of the pelvis is messing things up.
Run the InvDyn tool on it. Let's see what sort of moments are being generated on the pelvis by the board. If you don't know how to do this, it works just like the SOpt Tool (IK result and External Loads file), only no reserve actuators. It will run (if it doesn't, you have bigger problems!), filter the kinematics with <=6Hz too.
I'd also like to see your reserve actuator file (along with the InvDyn results).
Ultimately, I think you have to figure out an alternative means of incorporating the board... perhaps attaching to the feet?
Jake Banks

the Inverse dynamics works just fine, i think.
When choosing the pelvis to connect the board to, i thought it had to be directly connected to the highst body in the parent child hierarchie, so if the board moves the whole body does too.
the Inverse Dynamics file is too big to bw uploaded here. Can i send i to you in any other way?
thanks a lot.

Alex Aasmann

Can you attach the InvDyn results?
In addition, try increasing the optimal force(s) for all board and pelvis related Reserve actuators to 500. See if SOpt solves then. I think it will. As even during 'normal' situations the pelvis actuators need to be >100, as they help adjust things relative to the measured GRFs. Then your board joint reserve forces/torques absolutely need to high (or at least what the InvDyn results show you the moments/forces are), otherwise SOpt won't work because nothing else can hold it there (no muscles).

I hear you on your attachment strategy, but I think it messes things up. Even if the above works, it still won't be modeled correctly. Besides, if your tracking the person with the markers, they will move with the markers! I would suggest making a left and right 'board' (i.e., cut your board in half) and attach it to the bottom of the corresponding left/right foot. If you only have 1 GRF, then put that on the pelvis.

Jake Banks

here they are i had to split it in two

Yup... see in columns B-M??? those moments/forces have nothing to 'help' them in SOpt, and they are not trivial. So it is good to set your reserve actuator torque/force (in the RA file) for those moments/forces to be at least as high as the max you expect (probably 2x). Try that (in SOpt).

While your at it, you can make the Sternum Reserve actuators higher too. This joint is not physiologic in the model, make the 6DoF for that joint reserve actuators ~50, this helps things solve better too.

I don't think having a negative time hurts anything, but that could also be something to try and remove from your files, just in case it does (but since IK and ID run, I doubt it does).
Jake Banks

how do you know from the inverse dynamics file that i have to make my optimal force higher?
I have been setting the Optimal force at the RA file ato 500 and i still keep getting the same error, do you think the problem could lye somewhere else?

thank you