Inquiry about markers tracking error

Main Forum
1

Hi

I am using OpenCap to obtain the kinematics information which is given in .trc format. After converting this file to c3D format, I found that the markers are 16 units less than the AnyBody, which are all from the upper part body. For example, the markers annotated as red box are not found in my c3D data, so i eliminated those markers in MarkerProtocol.any .

c3dopencap
c3dopencap626×536 35.8 KB

The names of the markers between my c3D data and anybody were also quite different, so I manually changed the markers' names to match the AnyBody. I use the AnyBody sample application file - Plug-in-gait-Simple and using the main file named "FullBody_GRFPrediction.main" to run my motion capture data.

The errors as shown keep happening. It started from

Constraint #23 is above tolerance 0.001, error = 0.008946, constr. #2 in 'Main.HumanModel.BodyModel.Trunk.JointsLumbar.SpineRhythmDrv'.

to

Constraint #410 is above tolerance 0.001, error = 0.012513, segment constr. 'Main.HumanModel.BodyModel.Left.Leg.Seg.Patella'.

I changed the

#define BM_MANNEQUIN_DRIVER_DEFAULT ON

and the kinematics tolerance in the KinematicStudyForParameterIdentificationSettings.any from 1e-6 to 1e-3, the same issues happened. Finally, when I changed to 1e0 (although know it is the wrong setting), the errors left only a few lines as shown:

Progressing to solve kinematic optimality conditions and hard constraints.
Progressing to solve kinematic optimality conditions and hard constraints.
Progressing to solve kinematic optimality conditions and hard constraints.
Progressing to solve kinematic optimality conditions and hard constraints.
Failed to solve kinematic optimality conditions and hard constraints after 5 fallback attemps.
Constraint violations for study 'Main.Studies.ParameterIdentification.KinematicStudyForParameterIdentification' :
ERROR(OBJ.MCH.KIN3) : KinematicStudyForParameterIdentification.any(11) : KinematicStudyForParameterIdentification.InitialConditions : Kinematic analysis failed in time step 0 : Position analysis is not completed

May I know what the possible reasons for causing this to happen? Please advise
The picture below is how the model looks like at the moment it failed

anybody skeletal
anybody skeletal323×598 28 KB

2

Hi Mitchelle,

It looks like the marker protocol doesn't provide sufficient information for driving the full body model. Please be aware that using GRF prediction model without full body kinematics can lead to errors in GRF prediction. For example, you have no head markers and you would have to invent some motion for the head, which will most likely lead to errors in GRF prediction.

Is your model failing at the first time step? From the picture, it looks like all the markers from the c3d file are loaded at the origin. You may need to look into the settings that were used for exporting the c3d file. I suspect it could simply be a mismatch between units. Maybe the actual measurements are in meters but the c3d file tells that the units are in mm and it leads to this issue.

Hope this helps
Best regards,
Dave

3

Hi Mitchelle,

Regarding your problem with markers, it may be possible to fix this inside AnyBody. I am sorry, it didn't strike me earlier. If you are using the standard mocap model, please explore the model tree to Main.ModelSetup.C3DFileData.PointsScaleFactor, locate it in AnyScript and try to change this factor (for example, multiply the current value by 1000).

Best regards,
Dave

4

This topic was automatically closed 125 days after the last reply. New replies are no longer allowed.