I’m trying to run the GaitLowerExtremity model with personal C3D datas from vicon acquisitions. My C3D file contains a full body set of markers, as the “GaitNormal0003-processed” (from anybody repository) does.
My problem is that all the markers of the 2 legs connect to the ones in the C3D file, but the 4 markers of the head (RBHD, LBHD, RFHD, LFHD), and the 3 markers of the trunk (STERN, CLAV, T10) don’t want to connect with the model.
When I run the MotionAndParameterOptimization analysis I have the following error message:
ERROR(OBJ.MCH.KIN3) : C:/U…s/B…s/A…a/R…g/A…y/A…x/A…4/A…n/E…s/G…y/GaitLowerExtremity.main.any : KinematicStudyForParameterIdentification.InitialConditions : Kinematic analysis failed in time step 0
I would write something about the problems of your model.
※ About Kinematics(MotionAndParameterOptimizationModel process)
First, in your ‘Walking 3.c3d’ file, the positions of RANK & RHEE markers are strange.
So I could not run this model with this C3D file.
So, I tried to modify your model with ‘Jogging 3.c3d’ file.
When you would pass the ‘MotionAndParameterOptimizationModel’ step,
you may experience some kinematics errors.
In your C3D files, at early frames some of markers are missing.
So I had to postpone the beginning frame.
You can check this in the ‘TrialSpecificData.any’ file.
Also I modified several things in ‘ModelSetup.any’ file.
In ‘TrialSpecificData.any’ file, I could adjust the Anthropometrics data(segment lengths).
Also I could chage InitialPositionOfBody data.
You should see ‘Mannequin.any’ file if you want to know how these InitialPositionOfBody data are working.
In ‘ModelSetup.any’ file, I had to adjust the positions of markers.
In KinematicStudyForParameterIdentification and MotionOptimization studies,
I could increase the kinematic tolearnce values.
But please do not increase those values above 1e-3.
※ About Kinetics(InverseDynamicModel process)
You should set the right type of ForcePlate class.
It seems that you used Force Plate Type2(Maybe AMTI).
You should set the right gravity direction for correct inverse dynamics analysis.
When you see your model, the global normal direciton is ‘Z’.
So the gravity vector should be {0, 0, -9.81}.
So I defined this vector in ‘TrialSpecificData.any’ file.
And used this value in other studies.
It will take some time for you to get used to modify your model to pass the MotionAndParameterOptimizationModel well.
I hope that you can start your AnyBody modeling work well from this model.
Thank you very much for your help, now I am able to run the model with different C3D files I have (Jumping, Running and Walking)!
I understood the modifications you did concerning gravity, postponning the beginning frames and kinematic tolerance value but how do you know exactly how to modify the drivers and mannequin position? Is that only your experience or is there a technique that would enable me to do it myself?
It is good to hear from you that now you can run with different C3D files of yours.
The most difficult thing is to modify the position of markers, and to modify the initial values of joint angles(including hip positions and orientations).
And please keep in mind that modifying kinematic tolerance values are the last thing you have to do.
If you can avoid modifying kinematic tolerance, that’s the best way.
There is no special technique to do that.
You should rely on your eyesight and personal experience.
So we are trying to do our efforts to improve the usability of this kind of gait analysis application.
And maybe there will be some additional tips how to use this GaitFullBody model easily on our wiki page in future.
So please keep your eyes on our websites.
Best regards and hope you to do good modeling continually,
Moonki
