I have a question regarding using the c3d file (which contains both the marker position and the force plate data) and using GRF_prediction to predict the ground reaction force. Does the force plate measurement data from the c3d file complement the GRF prediction in Anybody or the GRF prediction using Anybody is completely independent of the force plate data? If the GRF_prediction does take into consideration the force plate data, how much of an effect is expected for the inverse dynamics analysis result when using the force plate and not using the force plate data (Does the muscle forces and the joint reaction forces differ in these two situations)? I am using Anybody 7.0 with AMMR 1.6.6.
It's either or. You can specify if you want to use the GRF prediction.
There are pros and cons to use both:
Human kinematics itself, and then modelled optimized kinematics, may be slightly inconsistent with the force plate data. In this scenario, it would be beneficial to use the GRF prediction case, which would minimize residuals in the system.
And conversely, you may want to intentionally rely on the true reading from force plates for your analysis, and then minimize system's kinetic residuals and trust optimized motion.
Depending on your goals and reliability of the data, you may want to opt for GRF prediction, i.e. in case of poor kinematics (frequently dropping out markers) or poor force plate readings (bad calibrations and so on) it may make sense to use the GRF prediction.