Static Optimization problem MoBL Arms Model

[Daniel Iordanov]

Hello, I have a problem with static optimization on MobL Arms Model.

The files I am using are:

- A motion file which represents a shoulder elevation (shoulder_elv) from 0 to 90, the other coordinates are all blocked because I am interested only in studying the effect on Deltoid and Latissimus Dorsi during arm elevation and only in the shoulder elevation coordinate.

- An external force setup in which I have defined a mass of 2.5 kg applied on the hand (center of mass),
simulating a tool that the subject has to lift during the arm elevation.

- An actuators file in which I have defined actuators for the 3 DOF of the shoulder (shoulder_elv, elv_angle,
shoulder_rot). I have set values of -inf and inf for control value and values of 1 for optimal force. I have not defined the rest of actuators for DOF because they are all blocked and they are not involved in the movement that I am representing.

After defining my files I perform Static Optimization and it seems to finish without failures but when I analyze the results I see 2 problems:

- During all the simulation the values for Performance at each step are too high which means that the model is struggling to find a solution, but there is no message of weak muscles.

- When I visualize the activations.sto in the plotter, no one of the muscles in the model has an activation higher than 0.07 and that is not real because physiologically, the deltoids and latissimus dorsi are the main muscles involved in arm elevation

By the way, for performing the simulation I have removed the CoordinateLimiteForce elements of the original MoBL Arms model.

Why I am not obtaining any activation for the muscles involved in the arm elevation? Is it something wrong in the model?
Should I define actuators for blocked DOF? Should I change the values for the defined actuators?

Any help will be appreciated, thank you in advance.

Best regards,
Dani

[Dimitar Stanev]

This model is a bit tricky to work with, due to its complexity, constraints and its use of the deprecated muscle model. You can try to run computed muscle controls, without appending any residual actuators. Maybe the results would be more reasonable.