I am trying to perform analysis using motion data acquired by a special set of force plates. As shown in the picture, there was one front force plate and one rear force plate. The red arrows are forces as recorded by the force platform. The blue arrow is the GRF associated with the foot. You’ll notice as a foot crosses between platforms, the blue arrow becomes the sum of the two red arrows. My question is: do I need to make changes in AnyBody to reflect such difference? All the models I am working on involves one FP specific to each foot with no foot cross-over.
I was lastly advised to look up “non-automated” foot contact detection. Unfortunately I haven’t been able to make much progress yet. Any hint on how to start will be greatly appreciated!
You should use the force plate class templates such as ForcePlateType2, ForcePlateType3 or ForcePlateType4 which also requires specific foot where the force will be transferred.
I am not sure if I have articulated my questions well enough. I understand I need to start from force plate class templates, in my case {4,4} or two type-4 FP.
Based on what I read (http://wiki.anyscript.org/index.php/GaitFullBody(GaitLowerExtremity) ), there are two templates available: ForcePlateType4AutoDetection or ForcePlateType4. However I am not sure either might suit my case, because one automatically detects which foot is in contact with which force plate, and the other does not detect contact since it’s known. For my case, since the force plates are arranged as front-rear, both feet WILL be in contact of BOTH force plates. I already tried ForcePlateType4AutoDetection, it didn’t give me error message but the calculated JRFs don’t look correct to me.
Another reason makes me believe that available templates wont work is that the GRFs measured in both plates are very different with typical cases (i.e., GRF of one FP is lagging behind the other, but both have similar patters such as two peaks at early and late stance).
Unfortunately, your motion capture trial is not a good case for our force plate class templates.
You may try to invent your new force plate class template, but it will be a little bit difficult to explain here because there should be some assumptions.
Is this a kind of very frequent case for your experiments?
If possible, I would like to avoid using this case.
Unfortunately I am just lucky to find this lab willing to provide motion capture data and I am not sure how easy (or difficult) for them to reset their experimental configuration. If I have to modify my FP template, is there any example for me to look at?
My collaborator calculated the GRFs experienced by both feet using Visual3D, which appears as normal GRFs measured by FP to me. These pseudo GRFs include 5777X3 matrix for each foot (frames by number of components). The real GRFs consists10 groups of 5777X3 matrix, with x-, y- and z- components for each group (5777 X 30 matrix in total).
My question is, is it possible to replace the GRFs measured by the nonconventional FPs with the pseudo GRFs measured for each foot? If it's possible we can use ForcePlateType4 to finish the problem.
I am a beginner in musculoskeletal analysis and your help is greatly appreciated.
I attached three files - one is my AnyBody model, which uses the c3d provided by my collaborator. I didn't change much except choosing a few frames. No error was detected by AnyBody but the results don't look right.
The other two txt files are the pseudo GRFs - each has only three (x-, y-, and z-) components. I didn't select any frame but I have confirmed the length of the pseudo GRFs are the same as the original GRFs.
Your help and suggestion will be greatly appreciated!
The pseudo GRFs will let you know the result forces on both left and right foot.
But it does not tell you the each component of the forces from both feet when both feet are on the same force plate.
And I realized that your case may come from the force plates which are installed on the treadmill.
You may be able to solve this issue by special assumption such as optimal distribution of the forces into both left and right foot. But that is beyond the scope of the help from this forum.