Not Obtaining New Mass Distribution When Scaling

[Stephen Riutta]

Hello,

Because of a subject's unique anthropometry, I would like to scale Sam Hamner's running model with a new mass distribution. I set the new subject mass and new scale factors with the scale tool (Version 3.0). However, the resulting scaled model has the same mass distribution as the original model regardless of whether the "Preserve mass distribution during scale" box is checked or unchecked. The geometry of the segments does scale correctly.

I've also tested this with large (non-physical) scale factors such as 1.5, and the mass distribution is still unchanged.

Does anyone else get a new mass distribution with new scale factors? If so, any ideas on what I'm doing wrong?

Thanks,

Stephen

[David John Saxby]

Hey,

May be a bug. Are you in v3.0 or the beta?
Either way, you can load the Osim model into a Mat structure and edit the inertia manually or through a function. A bit annoying though.

DJS

[Stephen Riutta]

Hi,

Thanks for the response. I'm using the standard (non-beta) version of 3.0. I've also tried doing the scaling via the command prompt, but got the same results.

I've been modifying the mass and inertia through the XML file and through the GUI. I'm interested in modifying the mass and inertia through the Mat structure though. Do I obtain this through the API?

Regards,

Stephen

[Ayman Habib]

Thanks Stephen,

Indeed the bug has been found and fixed and will show in the next release. For now you'll need to edit the scaled model in the GUI and update mass properties. There're helper methods to get/setInertia using an "ArrayDouble" that you can use for now.

Best regards,
-Ayman

[David John Saxby]

Stephen,

If you go the link below you'll find code developed by Tim Dorn (maybe extended based on other work, don't know the legacy of the code).

https://simtk.org/project/xml/downloads ... kage_id851

These functions allow you to read in Osim models into Matlab and edit the components of the model, including features such as inertia. You can then write the Mat structures back to Osim and then bob's your uncle.

Best,

David

[Stephen Riutta]

Hello,

David, thanks for letting me know about that toolbox. I'll certainly try that out.

Ayman, I'm looking forward to trying out the scaling tool when the next version is released. It's great to hear the bug has been fixed.

Regards,

Stephen