actually I am preparing a motion capturing experiment. I intend to analyse with inverse dynamics the movement of the upper body (especially arm-shoulder-system) for certain manual assembly tasks. The subject will perform the task in a (static) standing posture and handle different weights. There will be no movement of the lower body.
To prepare my investigation I have already checked the wiki and found a good description how to set up the motion capturing analysis. I have learnt I need three kind of data to run an Anybody model with motion capturing data:
Motion Capture data (C3D-file)
Data from force platforms
Anthropometric data of the recorded subjects
Two question are still remaining:
Are the data from the force platform really necessary, when I only investigate the movements of the upper body?
Is it possible to apply different loads to the model afterwards?
If you are only interested in the movement of the upper body so if you want to use only the upper body segments, then you don’t need to apply the force plate data into your model.
But if your upper body (including the pelvis segment) has any kind of interactions with external environment, then those forces should be applied to your model.
Of course you can apply any kind of different load to your model afterwards.
in several post I have read how to setup up ground reaction forces. The best way is to use data from force plates. If such data are not available AnyBody will calculate them. How can I figure out these calculated ground reaction forces?
Which marker setup should we use to get the best results? Are there any experiences?
The plan right now is that we want to use the Plug-In-Gait marker setup, but I read about this one that it is probably limited valid in the upper body section. Is that correct?
I think this one is the easiest one to implement into AnyBody.
Which ForcePlateType do you advise? We wanted to use Type2.
Personally I think that the quality of simulation depends on the quality of the measurements.
Also I believe that plug-in-gait marker set will work well in general for your purpose.
If you have any other idea about your marker set then you can define it as you want.
Also we don’t have any special preference for force plate types.
Officially our MoCapModel supports the force type 2, 3 and 4.
But even you can develop your own force plate class template if you know the exact specification of your force plate.
One comment is that there are many users who use the combination of VICON and AMTI.
This means that they use the plug-in-gait marker set with the force plate type2 setting.
I made some good experiences with using two markers at the elbow (medial and lateral) instead of just one like in plug in gait. I experienced less problems with the rotation and gimbal lock.