"Unknown error"

Hello AnyBody users,

I am wondering if someone can help me understand what is going on and possibly test this by running on a separate machine. I am trying to use C3D data obtain from a capture with the MS Kinect sensor. I was able able to run initial conditions and kinematics at first and for no known reason I was not able to run again due to the error pasted below. After reloading the model and restarting the software I still get the same error.

I am using the AMMR1.4.1 repository. Also in the .zip folder is the ScalingLengthMassFat.any file that needs to replace the file by the same name in the Body/Scaling location. There is a small change to this file to adjust the shoulder width scaling.

Please let me know if you get same error and/or can offer help to fix it.

Thanks for your help!

Jeremy


  1. InitialConditions (Operation: Main.Studies.KinematicStudyForParameterIdentification.InitialConditions):
  2. …Design variables have been updated.
  3. …Load-time positions have been re-established.
    ERROR(SYS1) : Operation : Unknown error : ‘No Class Information’

Dear Jeremy,

I tried to load your model but I could see the following error:
ERROR(SCR.SCN1) : C:/W…E/[…]/A…4/M…s/S…t/G…2/GaitFullBody.main.any : Cannot open file : C:\WORKSPACE[AMMR]\AMMRV1.4\Body\AAUHuman\ToolBox\Mocap\CreateMarkerClassNoScale.any
Model loading skipped

So it seems that you’ve created your own ‘CreateMarkerClassNoScale.any’ file but you have put them into your local repository so we can’t load your model.

Could you put your own files in your model folder and try to upload your model again?

Best regards,
Moonki

Moonki,

Sorry - a couple include file were left in but are not needed. Please just out-comment the lines shown below, before the main folder.

Thanks,

Jeremy

////Class used for defining a marker without scaled placement guess
//#include “…AAUHuman/ToolBox/Mocap/CreateMarkerClassNoScale.any”
////Class used to construct a TekPad on finger segments
//#include “…dy/AAUHuman/ToolBox/Mocap/CreateTekPadForceClass.any”
////Class used to construct a forceplate of type4
//#include “…n/ToolBox/Mocap/ForcePlateType4AutoDetection.any”
//#include “…an/ToolBox/Mocap/ForcePlateType4.any”
////Class used for defining which antrhopometics measurements
//#include “…Mocap/OptimizeAnthropometricsOnOff.any”

Main = {

Hi Jeremy,

I have tried to load your model and discovered a number of things:

[ol]
[li]nstep is zero because the calculation in modelsetup sets it to zero…, i simply added a fixed number instead, this resulted otherwise in a NaN error for me, but please try make a new calculation for it.[/li][li]The kinematic tolerances are too high i think for all studies, pleasse try bring them back to what is in the gaitlowerextremity model[/li][li]There seem to be something going wrong with the legs in the model i failed to make it run kinematically with them in but i had no more time, it worked if i took the legs off. Do you need legs?[/li][/ol]Best regards
Søren

Soren,

Thanks again for your help.

  1. You are correct, I had the ‘round’ function on the inside vs. the outside so it was either rounding to 0 or 1 depending one the tStart and tStop time. The line in ModelSetup was changed to the following and works correctly:

AnyIntVar nStep = round(30*(.TrialSpecificData.tEnd-.TrialSpecificData.tStart));

  1. I agree that these were high. I was just trying something/anything to get initial conditions to solve. If that’s not the biggest issue I will return the values to normal and focus elsewhere :slight_smile:

  2. I do need the legs. I also tried without the legs and it ran perfectly. With the legs the pelvis and legs do not following the markers and make an odd twisting motion. It is almost like a driver got switched on and took over… I realize you don’t have much time but even if you can just suggest I look somewhere to debug, I would appreciate it. I will keep trying but not sure I have a good idea at this point.

Jeremy

Soren,

Please disregard previous problem. In the joints and driver file I out-commented all but the arm drivers which where controlling pronation of the forearm and it worked. My general explanation is that I had too many drivers turned on. Maybe you can offer a better explanation?

Thanks again!

Jeremy

FYI: Here is what the file looked like:

AnyFolder Drivers = {
#include “…/…/…/Body/AAUHuman/BodyModels/include/MannequinDrivers.any”
// Aliases for convenient referencing
AnyFolder &JntPos=…Mannequin.Posture;
AnyFolder &JntVel=…Mannequin.PostureVel;

AnyFolder &RefHM = Main.Studies.KinematicStudyForParameterIdentification.HumanModel;
// changed JC 9-19-12

// AnyFolder &Thorax = .HumanModel.Trunk.SegmentsThorax.ThoraxSeg;
// AnyFolder &ref=.HumanModel.Trunk;

// The body parts are included only if there is no unhallowed combination detected
#ifdef UNHALLOWED_COMBINATION
#else

// #if TRUNK == 1
// #include “TrunkDrivers.any”
// #else
// #endif
//
//
// #if TRUNK_NECK == 1
// #include “TrunkNeckDrivers.any”
// #else
// #endif
//
//
#if RIGHT_ARM == 1
#include “RightArmDrivers.any”
#if RIGHT_DETAILLED_HAND == 1
#include “RightHandDrivers.any”
#endif
#else
#endif

#if LEFT_ARM == 1
#include “LeftArmDrivers.any”
#if LEFT_DETAILLED_HAND == 1
#include “LeftHandDrivers.any”
#endif
#else
#endif
//
//
// #if RIGHT_LEG == 1
// #include “RightLegDrivers.any”
// #else
// #endif
//
//
// #if LEFT_LEG == 1
// #include “LeftLegDrivers.any”
// #else
// #endif
//
//
// #if RIGHT_LEG_TD == 1
// #include “RightLegTDDrivers.any”
// #else
// #endif
//
//
// #if LEFT_LEG_TD == 1
// #include “LeftLegTDDrivers.any”
// #else
// #endif

#endif
};//Drivers

Hi Jeremy,

That makes perfect sense, I just did the same corrections and it seems run smoothly and converges…

Basically there was hard drivers applied to the legs with some “arbitary motion” and then at the same time it had to follow the markers which made it do the twist of pelvis, i also saw this.

It also makes sense to keep the pronation driver in there… with the markers configuration you have this dof can not be controlled, but by adding more markers that would be this possible.

Best regards
Søren