Analyzing arm curl motion

Main Forum
1

Hi, all

I’m trying to analyze arm curl motion with loading 10N downward(y-axis) on palm joint node. I used FreePostureMove model of repository7 and imported joint angle data related to time. I loaded the model successfully, but I failed to inverse dynamic analysis. I have no idea what is wrong with my model since it’s very simple motion. So, I really appreciate if you could check my model and provide me some advice.
And there is another question. After I run inverse dynamic analysis, I can check the output data in chartFX 2D. But if I want to check it again next time, I have to run the inverse dynamic analysis again. Is there anything I can do to see result of the inverse dynamic analysis anytime after I finished analysis?

Thank you!

Kind Regards,
Sungwoo Park

2

Hi Sung-Woo,
could you please describe the error in more detail? Which wersions (Rep and AMS) are you running. I loaded the model and it worked. Some remarks: Please be aware that the model you are using is grounded in in the pelvis. You can also exclude a lot of muscles to speed it up a little.
Basically it is not possible to store results in the System right now, but there is a way around. Just define a outputfile containing all information you are interested in and use e.g. excel to look at it.
Hope this helps.

Best regards,
Sebastian

3

Hi, Sebastian,

Wasn’t there any problem? That’s weird. I keep failing to analyze it.
The error message was like this :

ERROR(OBJ.LIB.OOS1) : D:/sung/AnyBody/R…7/Body/AA…an/Arm/Muscle.any : subscapularis_1.SPLine : Unexpected exception in the library OOSol :
OOSol exception : MISSING MESSAGE.

The inverse dynamic analysis was stoped during the analysis.

Thanks for your help.

Sungwoo Park

4

Hi Sungwoo,

I increased the time steps and could replicate similar errors. Basically it seems to be a problem with the wrapping muscles. Try to decrease the time step or slightly increase the SPLine.StringMesh of the muscle that fails. Don’t increase it way too much, this will slow down the analysis quite a bit. Please search also the forum for “wrapping” or “Spline”

Best regards,
Sebastian

5

Hi, Sebastian,

I decreased time step, and I succeeded in analysis. But the muscle forces and activities of some muscles of the shoulder & arm were little bit unreasonable for me. For example, Muscle forces of Teres Major and Pectoralis Major were almost e-018 scale, it’s very very small. What do you think?

Thank you.

Best Regards,
sungwoo Park

6

Hi Sungwoo,
basically a value of e-18 is of course only “numerical rubbish” so no muscle acitvation is necessary at this point.
Best regards,
Sebastian

7

Hi, Sebastian,

While analyzing model, I got another error message.

ERROR(OBJ.MCH.MUS4) : D:/sung/A…y/R…7/A…n/E…s/d…l/Dumbell.Main.any : Study.InverseDynamics : Muscle recruitment solver : solver aborted due to singular KKT matrix

What is singular KKT matrix? Is it problem with Recruitemet Solver?

Best Regards,
Sungwoo Park

8

Bonjour!
Yes it should be a problem with the recruitment solver. I’m sorry to say that I can’t help you solve the problem but just for fun you can check what wikipedia say on the subject of Lagrange multipliers and Karush-Kuhn-Tucker…

http://en.wikipedia.org/wiki/Karush-Kuhn-Tucker_conditions

/jokke

9

Hi Sungwoo,

I hope you had fun reading the wikipedia article :). So yes it means that the muscle recruitment solver could not find a suitable solution. This can be due to wrong boundary conditions, lack of reaction forces, muscles unable to carry certain dof or so. But i can’t know which of those without the model.

So please try to look for such suspicious things in the model. If you can’t find anything then upload it.

Best regards, Sylvain.