Large knee flexion and large GRF motion (squat)

[Hoon Kim]

Hello.

I am Hoon.
I am using the OpenSim with athletic movements to prevent sports injuries and improve performances.

I am currently trying to run RRA and SO with squat exercises using gait 2392 model. In this movement, weights were applied to subject with a weight bar.
So, there are differences compared to walking trials.
1. Ground reaction forces are much larger in this squat exercises than GRFs in walking trials because of barbell's weights.
2. Knee flexion ROM was much larger than walking trials.

I tried to run RRA first. I used the example RRA actuator and task files in the OpenSim-model-2392 directory. After click run, the whole body was moving up-and-down and side-to-side in a view window. I want to know why it happens and how I can keep the position of the body.

After that, I tried to run SO with an inverse kinematics mot file as an input. I used the example CMC actuator file in the OpenSim-model-2392 directory. Although the SO is working, I can see some messages many times during the process. The messages was shown at the time of large knee flexion. Below is that messages.

The model appears too weak for static optimization.
Try increasing the strength and/or range of the following force(s):
semimem_r approaching upper bound of 1
semiten_r approaching upper bound of 1
bifemlh_r approaching upper bound of 1
bifemsh_r approaching upper bound of 1
add_brev_r approaching upper bound of 1
add_mag1_r approaching upper bound of 1
add_mag2_r approaching upper bound of 1
add_mag3_r approaching upper bound of 1
glut_max2_r approaching upper bound of 1
glut_max3_r approaching upper bound of 1
quad_fem_r approaching upper bound of 1
gem_r approaching upper bound of 1
rect_fem_r approaching upper bound of 1
vas_med_r approaching upper bound of 1
vas_int_r approaching upper bound of 1
vas_lat_r approaching upper bound of 1
semimem_l approaching upper bound of 1
semiten_l approaching upper bound of 1
bifemlh_l approaching upper bound of 1
add_brev_l approaching upper bound of 1
add_mag1_l approaching upper bound of 1
add_mag2_l approaching upper bound of 1
add_mag3_l approaching upper bound of 1
pect_l approaching upper bound of 1
glut_max2_l approaching upper bound of 1
glut_max3_l approaching upper bound of 1
quad_fem_l approaching upper bound of 1
gem_l approaching upper bound of 1
rect_fem_l approaching upper bound of 1
vas_med_l approaching upper bound of 1
vas_int_l approaching upper bound of 1
vas_lat_l approaching upper bound of 1

I think I should modify something in actuator or muscle parameter or etc. I am not sure which part I should modify to run RRA and SO correctly for the squat movements.
I looked at the OpenSim documents websites. But I am not fully understanding how and which parts I have to change to run RRA and SO correctly, especially, in squat exercises with weights (large muscle force required and large knee flexion).
In terms of max isometric force for each muscle, I am worried about passive forces change, if I increase the max isometric forces for each muscle in the model.

Any response and advice would be very helpful for me.

Thank you very much for reading my question.


Sincerely,

Hoon Kim

[Thomas Uchida]

Are you certain the 2392 model is appropriate for simulating this activity? According to the "Musculoskeletal Models" page in the Confluence documentation (https://simtk-confluence.stanford.edu:8 ... tal+Models), the 2392 model "may be inaccurate during motions with high degrees of knee flexion". You may want to investigate Adrian Lai's model and his corresponding publication: "A musculoskeletal model for simulations that involve high hip and knee flexion", https://simtk.org/projects/model-high-flex/.

[Hoon Kim]

Hello. Tom and everyone.
Thank you very much for your response.
I looked at the model by Adrian Lai and I agreed with your recommendation.
I tried to run the process with the model.

While I operated RRA, the average residuals are too high, especially in pelvis_ty and pelvis_tx. And the whole model is still moving up-and-down, and side-to-side.
I searched similar problems in the forum posts. After searching, I think the problem occurs from an absence of a bar in the model.

The bar has barbells and the weight is over 40 kg.

These are my questions.

1. Should I make the bar in the model? (If yes, I would like to know reasons, too.)

2. If I need to make the bar in the model, how can I determine COM in the model?

3. Can I run RRA and SO without making the bar in the model?

4. If the bar is not needed in the model, how can I fix current problems?

Thank you very much for any response, again.

Sincerely,

Hoon Kim

[Christine Dailey]

Are you certain the 2392 model is appropriate for simulating this activity? According to the "Musculoskeletal Models" page in the Confluence documentation (https://simtk-confluence.stanford.edu:8 ... tal+Models), the 2392 model "may be inaccurate during motions with high degrees of knee flexion". You may want to investigate Adrian Lai's model and his corresponding publication: "A musculoskeletal model for simulations that involve high hip and knee flexion", https://simtk.org/projects/model-high-flex/.

Where can I download LaiArnold2017 model update? I clicked your link (thank you for this!) but I dont see the download option

[Thomas Uchida]

Where can I download LaiArnold2017 model update? I clicked your link (thank you for this!) but I dont see the download option

- Go to the simtk project page: https://simtk.org/projects/model-high-flex
- Click "Downloads" at the top
- Click "Previous Releases"
- Click "Download Package" under "UpdatedRefinedModel"

[Christine Dailey]

Where can I download LaiArnold2017 model update? I clicked your link (thank you for this!) but I dont see the download option

- Go to the simtk project page: https://simtk.org/projects/model-high-flex
- Click "Downloads" at the top
- Click "Previous Releases"
- Click "Download Package" under "UpdatedRefinedModel"

Thank you SO much!