simtk.org

Hi
I'm trying to do "Tutorial 3 - Scaling, Inverse Kinematics, and Inverse".
I got scale results below, that seems satisfactory:
total squared error = 0.0137265, marker error: RMS=0.0187606, max=0.036696 (R.Thigh.Upper)

but when running IK tool I get the results below:
total squared error = 0.0132869, marker error: RMS=0.0207029, max=0.076671 (R.Acromium)

Although the RMS error seems ok, but max marker error for Acromium seems too much!

I have tried changing weights for IK but didn't reduce the max error significantly.

1. Is it with my scaling?
2. Is it impossible to get better IK results?
3. Is this IK result good enough to pass IK process and start Inverse dynamics process?
4. if not how can I fix it?

For your information, I have read the documentation.

Best Regards.

Hi Yunus,

It is totally fine to get high max errors for this marker even if the scaling is done properly. The reason is that the torso is modeled as a single rigid body, whereas in reality the spine and the upper limbs have many degrees of freedom. If you try to track this marker better by increasing the weight, you will probably increase the RMS error.

Is there any tolerable range for scaling and IK?