Static optimization

[rajat das]

Hi,

I am running Static Optimization result. Kindly see the "message window result" -- "performance and constraint violation". Is it correct ? Kindly see the attachment.

[Dimitar Stanev]

It seems that the performance is quite high. This means that the muscles are strongly activated. What activation exponent do you use in your SO setup? The constraint violation is low which is good. You can also check if the results are correct by inspecting the activation of the muscles. If activation saturates to 1 there may be a problem, e.g. you need stronger or more muscles or other things.

Best

[rajat das]

Thanks for the reply.
Kindly suggest what is the best possible setting.

[Dimitar Stanev]

Seems good. Just make sure that the end time is correct and that you have added external loads if any. A note to your previous figure: it seems that the feet body is very small, this can be caused by scaling.

[rajat das]

Thank u. Yes my joint marker data are noisy and also I have less no of marker.

1. From the "subject01_StaticOptimization_activation.sto" file we will get the amount of muscle activation with respect to time ? Is these values are normalized means always it will be 0-1 range ?
2. From the "subject01_StaticOptimization_force.sto" file we will get the muscle force and moments in N and Nm. Is this understanding correct ?
3. In these both the files after all the muscle column there are column related to Inverse kinematics file , what these mean ?
4. In "subject01_StaticOptimization_force.sto" file there are few column towards end related to "calcn Fx, Px, Tx? Kindly tell me what these column means.

[Dimitar Stanev]

1) Yes.
2) Yes. ForceReporter reports any force and torque that is applied on the model.
3) No IK data (kinematics) are reported in these two files.
4) these are the external ground reaction forces, point of application and torque respectively.

[rajat das]

Hi Dimitar thank you for the reply.

I have some other doubts, request you to kindly clarify.

1. From the "subject01_StaticOptimization_activation.sto" file "FX, FY, FZ, MX, MY, MZ, hip_flexion_r, hip_adduction_r, hip_rotation_r, knee_angle_r, ankle_angle_r...", what these column means ?

2. From the "subject01_StaticOptimization_force.sto" file "FX, FY, FZ, MX, MY, MZ, hip_flexion_r, hip_adduction_r, hip_rotation_r, knee_angle_r, ankle_angle_r...", what these column means ?

3. My 3D coordinate file is related to Single Limb Standing (SLS). In opensim tutorials I have seen some platform below the model. How to create that? And what is the use of that ?

[Dimitar Stanev]

1,2) The pelvis is a free body it has 3 translational dofs and 3 rotational, so you have 3 moments (M_) and 3 forces (F_). The rest are the joint torques that correspond to the specific degree since the rest of the joints are pure rotational.

3) This is a body that is welded in the ground. The body is used to estimate the ground reaction force with the feet during forward dynamics. I would recommend not to use this for walking. I think there is a tutorial somewhere in the user manual.

http://simtk-confluence.stanford.edu:80 ... %27s+Guide

[rajat das]

Thanks.
When I was running SO or RRA during simulation my model is going down, which should not be the case. Required your suggestion or feedback.

By iteration do we need to reduce both Reserve actuators and Residual actuators optimal force ?? Is this is the goal ? Kindly explain what is the logic.

[Dimitar Stanev]

When I was running SO or RRA during simulation my model is going down, which should not be the case.

If the model is going down, you are not supplying the external ground reaction forces. Since the pelvis can be actuated through reserve actuators, large forces will be required to keep the model "floating" (superman).

As I said don't try to use collision as a mean for estimation of the ground reaction forces, since it's very unstable and required special handling, something that is not supported yet.

By iteration do we need to reduce both Reserve actuators and Residual actuators optimal force ?? Is this is the goal ?

You can inspect the residual forces after the simulation. If they are large, try to find the cause and correct it. The optimal force is used to permit larger reserve forces, but it does not solve the root of the problem.