Kinematic indeterminate errors when a body mass is changed.


I drive an upper body model to obtain joint moments, and recently have one problem (I use a model in the repository with a few minor changes).

For a few participants’ anthropometric data, the model fails, giving “Model is kinematically indeterminate” errors.

It turns out that the model works fine if a body mass is reduced by several kg.
(I didn’t try to change a height, though)

Is it likely that a certain ratio of the body mass and the height can make the model kinematically indeterminate?
If so, is there a simple way to get around this issue?
If not, what options do I have with this issue?


Hi Sunwook,

Does it happens for small and heavy models, with short and wide thorax? If yes then the problem comes most probably from an ellipsoid fitted to the thorax. This ellipsoid is used as gliding plane for the scapula. As it is scalled together with the thorax, it appears that for a very short and wide thorax the long axis of the ellipsoid becomes smaller than one of the short axis.
This causes the kinematic analysis to fail. It should be solved in a future release of the repository. But for now i think you just have to avoid some extreme cases if you can.

Best regards, Sylvain.

Are you sure that you didn’t change anything else than the mass-property? The kinematics should not care about the mass. Please re-check what you have done.
I’m not surprised that the kinematics fail when changing the anthropometrics. Usually, you need to alter the kinematics somewhat in order for the motion to “fit” the body. Of course it all depends on how you drive/constrain the model.

Hello Sylvain,

Thanks for the reply.

That issue happens with few participants’ anthropometric data. Their data seem to fall under the category you mentioned.
I hope that issue is addressed soon.


Hi Sunwook,

So it seems it is definitly the ellipsoid problem. Thank you for spotting it. We will fix it as soon as we can in one of the future release.

Best Regards, Sylvain.