I got a negative value for activation!

Provide easy-to-use, extensible software for modeling, simulating, controlling, and analyzing the neuromusculoskeletal system.
POST REPLY
User avatar
Yokoya Ryusuke
Posts: 8
Joined: Thu Apr 27, 2023 10:52 pm

I got a negative value for activation!

Post by Yokoya Ryusuke » Fri Nov 03, 2023 7:01 am

I ran a simulation with the forward dynamics tool and in the results there are negative values for activation.
The simulation conditions are as follows

The model "arm26" was used.
The simulation was performed for 20 seconds.
The excitation was entered only for "BIClong" and its value was increased proportionally from 0 to 0.12 from the beginning of the simulation (0 sec) to the end time (20 sec).
The model was set up so that no excitation was given to the muscles other than "BIClong" and their values were always constant at 0.
The initial state of the model was set to the value obtained when the simulation was performed without any input to the model (i.e., all muscle excitation was set to 0) and the model was in a state of complete weakness. Specifically, the elbow joint angle was set to approximately 23.2 degrees.

As a result of this simulation, the value of the "BIClong" activation, to which the input was given, decreased to a negative value immediately after the start and increased according to the input after a few seconds.
For the muscles other than BIClong, which were not given any input, the activation values decreased to negative values from the start of the simulation until the end of the simulation.

Shouldn't the value of activation take a value between 0 and 1 and not go negative?

I have attached the xml file for the control of the excitation input I used and the resulting file.Image[/img]
Attachments
result.png
result.png (116.71 KiB) Viewed 733 times
arm26_states_degrees.mot
(640.73 KiB) Downloaded 35 times
controls.xml
(5.88 KiB) Downloaded 28 times

Tags:

User avatar
Thomas Uchida
Posts: 1793
Joined: Wed May 16, 2012 11:40 am

Re: I got a negative value for activation!

Post by Thomas Uchida » Fri Nov 03, 2023 1:17 pm

The "arm26" model uses the Thelen2003Muscle model, which contains a numerical singularity at activation = 0. This may be the source of the unexpected numerical results. Some additional information can be found in the "Muscle Model Theory and Publications" section of the User's Guide (https://simtk-confluence.stanford.edu:8 ... blications) and in the API documentation for the Thelen2003Muscle class (https://simtk.org/api_docs/opensim/api_ ... ml#details) and the MuscleFirstOrderActivationDynamicModel class (https://simtk.org/api_docs/opensim/api_ ... ml#details).

User avatar
Yokoya Ryusuke
Posts: 8
Joined: Thu Apr 27, 2023 10:52 pm

Re: I got a negative value for activation!

Post by Yokoya Ryusuke » Tue Nov 07, 2023 1:19 am

Hi, Mr Thomas Uchida.
Thank you for your response.
My question has been resolved!

User avatar
shanlin wang
Posts: 1
Joined: Sat Jun 24, 2023 6:38 pm

Re: I got a negative value for activation!

Post by shanlin wang » Thu Aug 15, 2024 8:15 pm

Hi, Mr Yokoya Ryusuke.
May I ask how you solved it?I used forward dynamics and the muscle activation also appeared negative.
thank you!

POST REPLY