Inverse Dynamics top-down approach

[Alexander Ruf]

Hey everyone,

I'm a master student of the German Sports University Cologne and writing my master thesis at the moment. My project is about the kinematic and kinetic differences of squatting with 80% 1RM with weightlifting shoes and barefoot until muscle failure. I use OpenSim for my data analysis and I encountered a problem which I didn't manage to solve so far. One of my outcome parameters is the lumbar extension moment, that I want to compare with and without weightlifting shoes. It was very easy and intuitive to calculate the knee moments with a standard inverse dynamics bottom-up approach but as far as I understand, OpenSim does not support a top-down approach in the user interface, including a external force like a barbell.

My plan was to create a .sto file that includes the relevant information of the external load (the barbell) and its orientation in space and a .xml file which refers to this information and tells OpenSim that the point of force applications is at the torso and not on the calcaneus, like it is when calculating the knee moments.

Unfortunately, I ran into the same error message again and again and not even with the help of ChatGPT, the problem could be solved. The key content of the error message looks as follows: InverseDynamicsTool Failed: ExternalForce: No Data source has been set. I don't understand this, because I made sure that the .sto file is linked correctly to the external force in my configuration.

Does anybody have experience with an analysis like mine or maybe knows a different approach to this problem? I can't find anything online regarding this topic. I am using OpenSim 4.5.

Thanks and greetings from Germany!
Alex

[Nicos Haralabidis]

Hello Alexander,

Here are a couple of articles that fairly recently touched upon the points you raised in your post (hopefully they help!):
[1] https://link.springer.com/article/10.10 ... 21-09808-7
[2] https://www.sciencedirect.com/science/a ... 9024003087
[3] https://link.springer.com/article/10.11 ... 017-0077-x

Thanks,

Nicos Haralabidis

[Alexander Ruf]

Hey Nicos,

thanks a lot for your recommendations, I will have a lot at them!

Greetings,
Alex