Forums Anyscript.org  

Go Back   Forums Anyscript.org > Professional Forum > Debug model

Notices

Reply
 
Thread Tools Display Modes
  #1  
Old 04-10-2015, 10:35 PM
jcho44 jcho44 is offline
Junior Member
 
Join Date: Mar 2015
Posts: 9
Default Synchronizing C3D file to MoCap_LowerBody

Hello,

We have had some issues trying to input our C3D file that we converted from a program called the Motion Monitor. We have been in contact with the forums, but unfortunately we have not heard back from the person we were in contact with. We have resolved our issue with the C3D file and it seemed that we were not exporting the file correctly due to incorrectly naming each marker. We were able to change the code around a little to better model our C3D file, we altered the MoCap_LowerBody.main file. I attached our workspace that we used and this is from the initial LowerLegModel that we downloaded from the website. We were able to load our model with the new attached C3D file (this file is attached), however we are receiving an error when we try to run the actual model. We are able to see a model but it does not produce any results or movements. We have attached a picture of the error so you can take a look at that and any help on how to resolve this issue would be greatly appreciated. We were also hoping someone could send us the Type 5 Force Plate code or input it into the script and then send it back to us because we had to turn the forceplate data off as it was giving us an error when we tried to load the model. We had to turn this off because the code was about the type 4 plate and we have the type 5. We are also wondering if the EMG collected data from our quad, hamstring, and calf is successfully transferred and represented because we are not seeing any muscles in the dynamic model. We believe it is due to the error we are receiving when trying to load the model with the InverseDynamic set to on, we also attached a screenshot of this as well.

For our final result, we need a musculoskeletal model of the lower body, in which we can derive force plate data, EMG data, and a moving model. Potentially we would like to perform a finite element analysis. We hope someone can propose a solution to our issue or let us know if this is possible.

Thank you very much for your help,

John
Workspace.zip

FinalPresentation.zip

Error Message InverseDynamic.jpg

Error Message AnyBody Tech 4-2-2015.jpg
Reply With Quote
  #2  
Old 04-13-2015, 09:32 AM
melund2 melund2 is offline
Member
 
Join Date: Aug 2009
Posts: 69
Default

Hi John

Using the model with your own data (lab setup, marker protocol etc.) requires that you adapt the example models to your own data. That is not always a trivial task and unless you are using the exact same marker protocol and marker naming as the example the model will not work out of the box.

The error you are seeing is probably caused by either too many or too few drivers on some of the degrees of freedom. Also you need to make sure that you don't have any marker dropout in your C3D file. There are ways to handles small dropouts in AnyBody, but that is a little more advanced. So I recommend that you cut your c3d file to a section of 100 frames where your markers are all visible, and where the subject starts in a neutral standing position. That is good way to start. When the model work with that data, you can start to make it more complicated.

Regarding the type 5 force plate, then I have never seen data from that type before. I tried to load the c3d file with Mokka (open source c3d viewer), but it wasn't able to visualize the force either. I will try to ask around if we know anyone who have worked with that format before. If you know how to interpret the forceplate data, then you can implemt the type5 class template your self. It is however not a the easiest task if your are new to anyscript.

Finally, the reason you don't see any muscles on the model is not because the EMG data is not loaded. EMG data is not use as direct input to the model. The mocap model is really two models. A kinematic model (parameters optimization model) without muscles that finds unkown segment length, marker positions and outputs the joint angles. The second step is an inverse dynamics model, which have muscles. It use the scaling and motion from the first model to calculate the muscle forces etc.

I don't know if this helps you. Maybe you can find a way to convert the type5 data to something that is more common. Also try get familiar with the example model first, and work with some of the tutorials to understand how the model works.

/Morten
Reply With Quote
  #3  
Old 04-13-2015, 08:25 PM
jcho44 jcho44 is offline
Junior Member
 
Join Date: Mar 2015
Posts: 9
Default

Thank you for your response. I believe the problem may derive from how we create our model through the use of rigid bodies. These are clusters of markers and are necessary to make the model in the programs optitrack and the motion monitor. The results are then put into a c3d file and imported into anybody.

In terms of the force platform we are not well versed in anyscript and have only been using this program for a few weeks so personally writing the code is a mountainous task for us seeing as our project is due in two weeks.

M professor has asked us to look into the model further and know what force platform would be suitable for this study, such as brand and type. We also were wondering that using this specific model and the gait analysis given in the repository, if a finite element analysis is capable for this model and if we could obtain EMG results from specific muscle? Would it be possible to remove muscle from the model in order to obtain specif muscle results, or are they clustered per section.

Thank you for your time. Please let me know if you need more information about our specific needs.

Regards,
John
Reply With Quote
  #4  
Old 04-15-2015, 04:14 PM
melund2 melund2 is offline
Member
 
Join Date: Aug 2009
Posts: 69
Default

Quote:
I believe the problem may derive from how we create our model through the use of rigid bodies. These are clusters of markers and are necessary to make the model in the programs optitrack and the motion monitor. The results are then put into a c3d file and imported into anybody.
No. That is not the case. AnyBody only use the maker data and the force plate data in the c3d files.

Quote:
In terms of the force platform we are not well versed in anyscript and have only been using this program for a few weeks so personally writing the code is a mountainous task for us seeing as our project is due in two weeks.
No. You are right. That is probably not realistic. The AnyBody Modeling System can be quite difficult to use. The best background for a user is a cross between a mechanical engineer, computer scientist and doctor

Quote:
M professor has asked us to look into the model further and know what force platform would be suitable for this study, such as brand and type.
I don't have any good advise on force plate brands. But I suggest you use a force plate that provide output in a format that is common. That is usually type 2, type 3, or type 4 force plates. Take a look at this site .

Quote:
We also were wondering that using this specific model and the gait analysis given in the repository, if a finite element analysis is capable for this model and if we could obtain EMG results from specific muscle?
That is several questions in one go. You can configure any model to output forces etc. so that these forces can be used as input in your finite element software. The AnyBody Modeling System can not do the finite elements calculations for you. It is a multibody modelling framework that you can used to calculate the boundary conditions for your finite element software.

About the other question. The software doesn't give you the simulated EMG, but it can calculate the the muscle activation for any muscle in the model. EMG and muscle activation may not always be the same for a number of reasons.

Quote:
Would it be possible to remove muscle from the model in order to obtain specif muscle results, or are they clustered per section.
I am not sure I understand your question. It is possible to remove individual muscles from the model and investigate the effects.

/Morten
Reply With Quote
Reply

Tags
c3d, debug, forceplate, mocap_lowerleg, musculoskeletal modeling

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +2. The time now is 07:21 AM.