Active Fiber Force when Activation set to 0

[Axel Koussou]

Dear OpenSim experts,

I am using the Rajagopal's model and I noticed that when I plot the active fiber-force of a muscle after setting Activation override to 0, I get a non-zero active fiber-force :

241674846_152267227091066_6106137179652424090_n.png (90.24 KiB) Viewed 1315 times

How can you explain this observation ?

Thanks in advance,

Regards

[Axel Koussou]

Dear OpenSim experts,

Since I have not get any answer, I take the liberty to re-post :

How do you explain that the active fiber-force of a muscle is not zero, after setting Activation override to 0 ?

Thanks in advance,

Regards

[Thomas Uchida]

How do you explain that the active fiber-force of a muscle is not zero, after setting Activation override to 0 ?

My understanding is that the plotter follows different rules depending on the source of the data being plotted. I'm not sure what rules are being followed in this case, why the "Activation override" parameter is not being applied, or how you have generated this plot. One hypothesis is that the plot is being generated from a Storage that does not contain any activations, in which case there is nothing to override. The active fiber force should be zero if activation is zero, so I may also be misunderstanding what you are trying to calculate.

[Axel Koussou]

Hello Thomas,

Thank you for your return.

I'm interesting in getting the passive moment of the joint in function of the joint angle.
For that, for the ankle, I plot the relation between the ankle moment and the ankle angle, after having select all the ankle muscles, and setting the activation override to 0, as you can see on the image below.

Thus, I wanted to be sure that when I set the activation to 0, no more active fiber force were present. That's why I did this quick verification, as explained in my first post.
Since, I just load the model and plot the active fiber force, your hypothesis "that the plot is being generated from a Storage that does not contain any activations, in which case there is nothing to override", might be true.

But then, my question is : Does the passive moment that I plot is correct ?

Thanks in advance,

Regards

[Axel Koussou]

Dear OpenSim Experts,

I take the liberty to re-open this question since I did not have a complete answer.

As a reminder, I'm interesting in getting the passive moment of the joint in function of the joint angle.
For instance, for the ankle, I plot the relation between the ankle moment and the ankle angle, after having select all the ankle muscles, and setting the activation override to 0, as you can see in the attachments.

Nevertheless, I wanted to be sure that when I set the activation to 0, no more active fiber force were present. Unfortunately, I noticed that when I plot the active fiber-force of a muscle after setting Activation override to 0, I get a non-zero active fiber-force (see the figure in my first post) and I do not understand why. Should not be equal to 0 ? And is the way I use to obtain passive moment (setting activation to 0) is correct or not?

Thanks in advance,

Regards

P;S : You could notice some differences between my two images (name of the X-quantity for instance). It's just because since my last post I have modified my model a little bit but the phenomenon is still the same.

[Najoua Assila]

Hi,

I think while the activation was overridden to 0, Opensim relies on the default values (min activation) of the muscle to plot the active fibre force. I am not sure if you are using a Thelen muscle but as you can see here : https://github.com/opensim-org/opensim- ... Muscle.cpp, minimum activation can not be below 0.01 (due to the singularity at 0).

To confirm if this observation does indeed influence the figure output, I tried plotting the active fibre muscle with an activation set to zero in the plot too (Override activation to 0), with two different values of the min activation specified in the muscle properties, and indeed, I do find two significantly different curves.

Hope this helps,

Najoua

[Axel Koussou]

Hi Najoua,

Thank you for your return. Indeed, it helped me.

Actually, I am using a Millard muscle(I should have precised that the Rajagopal mode uses Millard Muscles in my first post), and with this kind of muscles minimal activation can be set to 0. Nevertheless, in the Rajagopal model, minimal activation of all the muscle is set to 0.01. Thus, I modified this properties to 0 and, as previously, I plot the active fiber-force of a muscle after setting Activation override to 0. This time, I obtained a zero active fiber-force.
Moreover, for the ankle passive moment-angle curve, I obtained little difference (few Nm) between the two conditions (minimal activation to 0 or to 0.01).

Thus, it appears that, as you stated, even if the activation is overridden to 0, Opensim relies on the default values (min activation) of the muscle to plot the active fibre force (and passive moment).

Thanks in advance,

Regards