I’m an italian student and I’m using Anybody 6.0.2 for my thesis.
When I open the human model and define detailed hand on (with the command #define BM_ARM_DETAILED_HAND_RIGHT CONST_HAND_DETAILED) i get a problem when starting initial conditions.
The error is:
“1) …Load-time positions have been re-established.
WARNING! NAN reached!
Unexpected error in computational kernel
Constraint no. #0 in ‘Main.HumanModel.BodyModel.Right.ShoulderArm.Mus.ArtificialRake.LinCom2Drv’ above error tolerance 0.000001, error = 0.100000.
ERROR(OBJ.MCH.KIN3) : C:/U…s/A…y/D…p/TESI/T…y/D…a/S…e/P…o/Prova_mano.main.any(50) : Study.InitialConditions : Kinematic analysis failed in time step 0”
It seems like default Joints and Drivers of the hand are missing.
You have answered your question - drivers are missing (or there are duplicates). Unfortunately, we do not have default drivers for the detailed hand in the repository.
I would recommend to start working with a model that includes detailed hand or copy the drivers from there. For example, FreePostureFullBodyShoulderRhythmStatic.
Thank you for your suggestion, it was very useful.
I tried copying the drivers but had problems with folder names. I also tried the FreePostureFullBodyShoulderRhythmStatic model but I was afraid it was incomplete for my purposes.
So I decided to manually copy and rewrite the drivers from that template into the default human model. It worked properly!
Hope I will have no problems when I start simulations.
From your description is it not easy to know what is wrong, to be able to help better please make a description of which reactions you have made between hand and phone.
I can think of these items:
[ol]
[li]The detailed hand model does not have the arm muscles attached to the fingers, instead there are torque muscles in the finger joints.
[/li]
[li]The constraints between hand and device could be wrong, how was this done? is the thumb pressing ?
[/li]
[li] What is the mass of the phone?
[/li]
[/ol]
That’s the point. I think the reactions are the problem.
The phone is constrained on the back with a spherical joint on a point of the distal phalanx 3. The thumb simply follows a traiectory on the screen.
Now I activated reactions in the spherical joint driver (they are On) a something seems to change in a positive way.
The mass of the phone is 133g (0.133 in anybody).
The model has only one arm (no legs and left arm) and it’s constrained on the pelvis.
Another constraint is a prismatic between the phone and the skull to fix the position between the eyes and the screen. This prismatic is guided to define the distance between the eyes and the screen.
The prismatic joint you have between head and phone is a add reactions you do not actually want i think… you essentially have the head to carry some reactions of the phone… You can disable these reaction in the prismatic joint by writing something like:
Constraints.Reaction.Type={Off,Off,Off,Off,Off};
Inside the prismatic joint this will disable reactions but keep the kinematics.
I think it is a good idea to have the reaction on the spherical joint driver, will be like the person is firmly holding the phone with the hand.
Which force do you have on the thumb is it pressing?
Thanks for the command “Constraints.Reaction.Type” suggestion.
I didn’t model the force on the thumb. It’s basically a kinematic movement The phone has a touchscreen panel and I supposed the absence of forces (no physical buttons to press).
Your suggestion has been very helpful and I was wrong.
I have to simulate a Force and I’m thinking how (to simulate tapping).
Sorry for my misunderstanding.
I have tried to use AnyForce3D (constant to start)
on the thumb but unfortunately
the activity doesn’t seem to change.
I’m using experimental measurements (EMG and %MVC).
I’m trying in this moment and using an enormous Force the model
muscle’s result overloaded.
The strange thing is that muscles like extensors and abductor of
the thumb show e^-10 order activity and they are the muscles I
want to study principally.
I think reactions are always the problem. I’m using reactions
on the tip of Finger3, Finger 4 (back of the phone).
I have to switch on reactions on the thumb’s tip driver?
I am just guessing now, the spherical joint connecting the phone which segments on the hand does it connect to?
I did not understand this with the thumb muscles (extensors and abductor)… the model has only torque generators for the finger joints the “real” muscles does not span the finger joints this is a simplification.