simtk.org

Dear all,

I have an OpenSim model where the limb length is of the order of 1-2mm. I also have the kinematic description of the model for a particular gait activity. My goal is to find out the musculoskeletal forces for the gait activity using the Inverse Dynamics principle.

Presently, I am trying to incorporate the motion in the OpenSim model through the Inverse Kinematics Tool. The RMS marker error in my model is around 0.0025m for the entire period while the maximum marker error is around 0.0026m. However, this small error of 2-3mm is relatively big for my model which has limb length of the order of 1-2mm. Consequently, the motion is not properly captured in the OpenSim model.

Under this scenario, I would like to know:
1) Is there any way to attain sub-millimeter marker error in OpenSim Inverse Kinematics module?

2) If the above is not possible, is there any alternative way to perform the Inverse Kinematics and Inverse Dynamics solution of my OpenSim model in order to calculate the musculoskeletal forces? Would it be a good idea to scale up the model and corresponding kinematics data (say by 20), perform the Inverse Kinematics and Inverse Dynamics solution in the scaled up version and finally scale down the musculoskeletal forces to the original model? Is there any published literature that correlates the scale of a model and the generated musculoskeletal force?

Any help in this regard would be highly appreciated.

With kind regards,
Kaushik

Is there any way to attain sub-millimeter marker error in OpenSim Inverse Kinematics module?

Have you tried setting the accuracy property of the InverseKinematicsTool? (To see the XML template for the InverseKinematicsTool, select "XML Browser" from the "Help" menu in the GUI, and select "InverseKinematicsTool" from the "Class Name" drop-down box.)

Thanks Tom. It worked. Thank you once again!
With kind regards,
Kaushik