simtk.org

Our goal is to perform a joint reaction analysis of the Vasavada Neck Model that was originally developed in SIMM. We are having trouble doing this and would be grateful for any insight that can be provided.

When using the 'Import SIMM Model' tool in OpenSim 3.3, the muscles from the .msl file are automatically converted to Schutte1993Muscle_Deprecated. However, our first step towards the goal of doing a joint reaction analysis is to run a forward simulation for 1.0s with all coordinates locked (creating a static .mot file). In this situation, using both the simulate button in the toolbar and the forward dynamics tool, we are getting an error that the 'Required condition 't1>t0' was not met. We have no additional constraints in our model. Because of this we were not able to move forward with the analysis.

I've also recreated the model by changing to Millard2012EquilibriumMuscle, and keeping everything else exactly the same, and in that situation, the simulation runs. However, we run into other issues using the Millard muscle and would much rather stick with the Schutte model, if possible.

Any suggestions would be much appreciated!

We don't support the use of Schutte muscles any more, hence the 'depreciated' in the name. Why not use the Millard type?

We tried to convert our muscles from Schutte model to the Millard model by copying in the appropriate muscle parameters. When we opened our models in OpenSim and checked the forces, they were extremely low. For example, one muscle with a maximum isometric force of approximately 170 N (and is close to optimal length in the neutral position), now had active fiber force of 8 N. With the Schutte model, the force in that muscle was 170 N. We also looked at Calvin Kuo's neck model, which uses the Millard model. It had a fiber force of 5 N in the neutral posture, when the max isometric force is 69 N.

Do you know why this might be happening, or do you have any suggestions for a better way to implement the Millard model?

Thanks!

For example, one muscle with a maximum isometric force of approximately 170 N (and is close to optimal length in the neutral position), now had active fiber force of 8 N

.

It is not clear if you have an issue. What was the activation and muscle velocity? Just because the max isometric force is 170N doesn't mean you should expect 170N at the optimal fiber length. Why would you assume the Millard muscle is wrong and Schutte model is correct?

It sounds like you should make a simple model, like the tug of war model distributed with OpenSim, and test the two muscles to get a better idea of how the muscles work.