Kinematics and Static optimization results

[Rahid Zaman]

Hello,
I am using Full body Lumbar spine model ( https://simtk.org/projects/fullbodylumbar ) in OpenSim 4.1. It showed some warnings about some joint definition (rotation and coupled), as the model is for OpenSim 3.3. So, I changed those joint definitions manually to coupled. Also, I convert the .osim file using the "Convert tool". In both cases, the warnings were gone.

The motion and GRF file are generated in another software simulation. But, after inserting the motion and GRF file, the model's motion is little bit different from the original. The model should bend forward more.I think it is mainly for flex_extension. As the model is for OpenSim 3.3, I tried with OpenSim 3.3 also. But, still showed the wrong motiono If I manually change any value from "Coordinates" tab, at that time it changes the color to red and is showing the original posture. But, when I am playing the motion again, it is showing the wrong motion. I can see the flex_extension is changing, but it seems that it is not contributing to the motion.

But, I have moved forward and run the Static Optimization. The Static optimization motion is actually similar to the original motion. But the problem is, there are some spike in Static optimization results, specially Latissima Dorsi and Internal Oblique muscle activation and forces. I mainly need that two muscles activation and forces. Also, there are lots of constraint violations during Static Optimization. But, it didn't mention "Static Optimization is FAILED" .

So my questions are:
1. Why is the model not showing original motion normally, but it is showing after Static optimization? Is this thing somehow contributing the spike in the Static Optimization output ?
2. How can I remove the spike in Static Optimization as well as the constraint violations.

I tried several ways to solve the problem:
There is no noise in kinematics and GRF file, as those were generated from simulation and filtered.
I converted the .osim file using the "Convert file" tool. It is still showing same results with spike.
To find out the reason about the constraint violations, I run the ID, but it is also showing the wrong motion.
I used the Reserve actuators given with the model and increases the force(Fx, Fy, Fz) and moment (Mx,My,Mz). But, still getting the spike.

Could any of you give me any hint about how can I solve these problems ?
Thanks in advance,
Rahid

[Carmichael Ong]

The spike in static optimization is very likely due to the issue with kinematics, since the tool calculates the inverse dynamics moments from the input kinematics. While it can be tempting to move on to the static optimization step to get results, you must make sure the input data (from inverse kinematics) is correct first. That model is extremely complicated, and the link you placed below mentioned that it was created for OpenSim 3.2. You may want to try to use OpenSim 3.2 first, or contact the authors to see if they have similar kinematics issues as you stated.

If you want to use OpenSim 4.0+, you may want to consider using a model that has been validated to work in these later versions. Depending on your research question, you also may be able to find a simpler model to better suit your needs.