Passive fiber force without exceeding optimal fiber length

[Adrian Lai]

Hi all,

I was just wondering if anyone has come across the problem of obtaining passive fiber forces without the normalised muscle fiber lengths exceeding optimal fiber length. The only way I feel this could happen would be that the normalised passive muscle fiber F-L curve has been skewed left such that passive forces can be developed prior to the optimal resting fiber length.

I'm obtaining these outputs (PassiveMuscleFiberForce and NormalizedFiberLength) by running MuscleAnalysis following my CMC simulations with a muscle model made up of predominantly Millard2012EquilibriumMuscle muscles. I've kept the FiberForceLengthCurve properties default (e.g. StrainAtZeroForce = 0, etc.).

I'm keen to hear if anyone else has encountered this problem and/or have a possible solution.

Thanks!

Adrian

[Dimitar Stanev]

Hi,

I was just wondering if anyone has come across the problem of obtaining passive fiber forces without the normalised muscle fiber lengths exceeding optimal fiber length.

Are you sure that those forces are passive? Maybe the muscle develops forces due to the constraint that the activation level can't be zero, so there is always an active force.

Best

[Adrian Lai]

Hi Jim,

Thanks for the reply. According to the outputs from muscle analysis using CMC (i.e. PassiveFiberForce and ActiveFiberForce), the muscle does develop active force (due to the constraints of the activation levels) but has passive forces as well so the accumulated force is shown in the ForceReporter outputs.

At the moment, I'm questioning the validity of partitioning of the passive and active outputs for a Millard muscle model based on the OpenSim documentation (https://simtk.org/api_docs/opensim/api_ ... uscle.html). It is possible these outputs may only work for previous muscle models but I'm not sure.

Adrian

[Thomas Uchida]

Hi Adrian,

It's difficult to investigate the issue without more information. What version of OpenSim are you using? How have you set the ignore_tendon_compliance, ignore_activation_dynamics, and fiber_damping flags? Uploading a minimal working example would also be helpful.

Tom

[Adrian Lai]

Hi Tom,

Thanks for the advice about the flags. I didn't know that the fiber damping value could affect the passive forces. After relaxing this value a bit, I was able to get rid of the passive forces (when the lengths were below the optimal fiber length) as well as negative passive force values (which were unrealistic).

I can't find much information about the fiber damping value. Where does it come from and why is it used? I thought it only affected the singularity problem in the force-velocity curve and possibly the oscillations in the system. It seems to have a significant influence on my simulations succeeding and failing as well as influencing the force predictions.

Thanks for your help.

Adrian

[Thomas Uchida]

Hi Adrian,

The Millard2012EquilibriumMuscle model is described in detail in the following publication:

Millard, M., Uchida, T., Seth, A., Delp, S.L. Flexing computational muscle: modeling and simulation of musculotendon dynamics. ASME Journal of Biomechanical Engineering 135(2):021005, 2013. http://nmbl.stanford.edu/flexing-comput ... -dynamics/

The paper should answer most of your questions. Note that the MuscleAnalysis includes the damping term when reporting the passive fiber force (i.e., it reports fiso*(fPE + beta*v) – see equation (8) in the paper).

Regards,
Tom

[Adrian Lai]

Oh perfect. Thanks Tom!

Adrian