Scaling of model while using own C3D file

[DEBADRATA]

Hi,
I am trying to run a C3D file of my own and I am using ANYBODY V 7.1. I could run the file by making adjustments of markers, drivers, etc. in the lower body model that comes by default. But now I want to know the exact method to adjust the mass distribution and segment lengths in the model. Please tell me where should I start from in my code of MoCap driven lower body model. I would like to have a method which automatically would take the scaling factors using the marker data in C3D files.

[toerholm]

Hi DEBADRATA,

Sorry for the slow reply, it is a good question.

In the file “SubjectSpecificData.any” you can set the masses of the individual segments.

e.g. like this

SegmentMasses ={
     
       Thorax =14;
     
     }; 

If you look in the ModelTree “Main.HumanModel.Anthropometrics.SegmentMasses” you can see all the values which can be set. By default a fixed percentage of the total body mass will be set for each segment.

If you want to do it automatically you will need make some equations for each segment mass based on its length, i do not have example code for this. One problem I could imagine would be that the model may end up with a total weight different from what you want, so it think you will need a correction factor to ensure total weight will be ok.

Best regards
Søren

[DEBADRATA]

Hi,
Thank you for your reply. I hope while using MoCap driven models, the body segment lengths are automatically taken as input using the marker data. The segment lengths that we can change in the script are only for an initial guess I think. Then by using the information regarding total mass of the body, the solver finds the mass of muscles, fats and bones using segment lengths too. Am I correct? Please rectify me if I have understood it wrong.

[toerholm]

Hi DEBADRATA,

This is correct the segment lengths are only used as initial guess, the parameter optimization will change them. The segment masses will not change during the parameter optimization, the segment masses are found using fixed ratios of total mass or by setting the values manually as i wrote in previous post.

It should be possible to make equations for the individual masses based on total-mass and segment lengths so that the masses would be changed by paramter optimization, but i do not have such an example.

Best regards
Søren

Best regards
Søren