I would like to get some clarification on wrist marker definition used in AMMR/Plug-in Gait model.
In my research I used Vicon equipment lab setup to record my mocap data and I followed their marker definition for Plug-in Gait for the upper limbs here.
However, when I loaded my C3D file onto AnyBody model, the wrists are always twisted.
The right wrist definition from AnyBody seems to follow as what have been defined from Vicon page but not for the left wrist. Even the right wrist is according to the definition plug-in gait, the forearm bones are twisted in Model View, the ulna and radius is crossing each other when it is not supposed to do that in neutral anatomical position.
Could someone point out to me where I have gone wrong from this? Because some of my simulations did fail but were successful if I turned off my arms. I am using AMMR v2.3.0
Is the image you posted before or after running the parameter optimization study?
It might be due to only having one marker at the elbow, you can try and switch on the default mannequin driver for that degree of freedom, it is only a soft driver but sometimes it can help the solver.
The image is after parameter optimisation; during 'run analysis' studies.
I agree with you due to only one marker for the elbow. I will try your suggestion with the soft driver. I didn't try it before because I thought I can only use the driver if any of the markers missing and worried it might break the simulation.