The gait2392 model has only one degree of freedom at the knee (knee flexion).
It is a common misconception that you need 3 rotational degrees of freedom if you want to study the 3D joint moment.
In fact, this would cause many problems. Inverse kinematics would become less reliable and sensitive to soft tissue artifact in the thigh markers. Static optimization would produce extremely large muscle force values, because knee muscles have a very small moment arm for internal/external rotation, thus requiring enormous muscle forces to produce even a small internal/external rotation moment.
Instead, keep the model the same, and use joint reaction analysis. This will give you a 3D moment and a 3D force, even when the joint has only one degree of freedom. See:
https://opensimconfluence.atlassian.net ... s+Analysis.
One thing to watch out for is that this 3D force and moment will be expressed in a cartesian coordinate system, attached to the ground, or the tibia, or the femur. It's NOT a joint coordinate system. In orthopedic joint mechanics research, they usually use the tibia coordinate system, for interpretation of the 3D joint load in terms of ACL injury risk, etc. See [1] for more information.
This is easier, and will produce more accurate results. You will essentially assume that the other degrees of freedom are controlled by passive forces (ligaments, articular contact), rather than muscles. This is not 100% correct, but much closer to the truth than making the muscles responsible for controlling all kinematic degrees of freedom.
It might be a good exercise to do the analysis both ways, and see the difference, but you can save yourself a lot of work if you trust my advice
Ton van den Bogert
[1] Derrick TR, et al. (2020) ISB recommendations on the reporting of intersegmental forces and moments during human motion analysis. J Biomech 99:109533.