Hi Soren,
We send you a simulation because we have the same error in a lot of
test, the kinematic simulation is correct but when we make the dinamic
analysis we have this error.
ERROR(OBJ1) : D:__AnyBody_LEMUS_Rep6\ARep\Aalborg_LEMUS_Model-
LEMUS.Main.any(93) : ‘Study’ : Muscle recruitment analysis
failed, simplex solver reports that solution does not satisfy all
constraints.
We send you a file (muscle error)with a simulation.
Best regards.
Hi Raul
I have tried to test your model a bit.
My conclusion is that it is mainly fluctuations in the accelerations
whcih could be the root of your recruitment problems. The error you
get means that the system could not find any way to recruit the
muscles and fullfilling the equilibrium equations.
If for example you look at the property rrdot of the hand (this is
the acceleration) you will see that it is very fluctuating and has a
size equivalent to gravity.
I have tried to increase the bSplineOrder in the JointsAndDrivers.any
file from 6 to 8 for all the drivers, this smooth accelerations a
bit.
In the main file i have tried to add a small amount of gravity in the
x and z direction so that the gravity vector now looks like this:
Gravity = {0.01, -9.81, 0.01};
We have previously found that this helps stabilizing ill conditioned
problems, and the amount of gravity added do not effect the results.
In the Study of the main file i have also outcommented the
kinematicTol=0.01; line, so that it now use the default value, the
tolerance of 0.01 is too high i belive, and may cause you problems,
in terms of fluctuations because the kinematics is not solved
accurately enough. I have also added the line
RecruitmentLpPenalty=0.000001; to the study, this will add a small
linear penalty to the muscle recruitment objective function.
These changes seems to make the model run a bit further than before
but it still do not run all steps. As mentioned above i think in this
case the main reason is the accelerations, if you have accelerations
of 6m/s^2 in the hands which are fluctuating between each timestep
and at the same time the load is applied in the hands i think the
results will not be very good.
We have however recently seen an example of a FreePostureMove based
model which had apparently unexplainable recruitment errors, which we
have tracked down to originate from the spine somehow. In the next
release of the repository we hope that this issue will be resolved,
but i hope smoothing the movement and thereby the accelerations will
be enough to solve your problem.
Please ask again if you hae further questions.
Best regards
Søren
— In anyscript@yahoogroups.com, Raúl Borraz Lamiel
<raulborrazlamiel@…> wrote:
>
> Hi Soren,
>
> We send you a simulation because we have the same error in a lot of
> test, the kinematic simulation is correct but when we make the
dinamic
> analysis we have this error.
>
> ERROR(OBJ1) : D:__AnyBody_LEMUS_Rep6\ARep\Aalborg_LEMUS_Model-
> LEMUS.Main.any(93) : ‘Study’ : Muscle recruitment analysis
> failed, simplex solver reports that solution does not satisfy all
> constraints.
>
> We send you a file (muscle error)with a simulation.
>
> Best regards.
>
Hi Soren,
I’m writing to you from the University of Zaragoza, Spain. Some months ago, my
work mate Raul asked you about an error that anybody returned on the dynamic
analysis. The error was and it continues being the following:
ERROR(OBJ1) :
D:__AnyBody_LEMUS_Rep6\ARep\Aalborg_LEMUS_Model-LEMUS.Main.any(93) : ‘Study’
: Muscle recruitment analysis failed, simplex solver reports that solution does
not satisfy all constraints.
We are trying to export real movements from a Motion Capture Software and we
are using the FullBodyModel BodyModel.any. Anybody runs the kinematics analysis
and completes it perfectly. The kinematics constraints are rightly resolved
but the inverse dynamics analysis doesn’t work.
We have applied your previous suggestions but it continues failing in simple
and slowly movements like lifting the right hand and then going it down. It’s
very strange because sometimes when we generate a movement in Motion Capture,
anybody works well, but it can fail in a similar movement.
I have attached all files we are using in the model, a video with the movement
in the Motion Capture application, the photos generated in anybody while the
kinematics analysis works well, a text file which contains the error, the files
that have been changed from the original inside Aalborg folder, and finally
inside _LEMUS there are the files generated by MotionCapture. Maybe this helps
you to find the mistake.
We have to give in some studies but this error doesn’t allow us to finish them.
Please, If you have any idea about what’s wrong with that I would be grateful
for your help.
Thank you very much.
Isabel Garcia
Boardwalk for $500? In 2007? Ha!
Play Monopoly Here and Now (it’s updated for today’s economy) at Yahoo! Games.
Don’t pick lemons.
See all the new 2007 cars at Yahoo! Autos.
[Non-text portions of this message have been removed]
Hello Isabel,
The muscle recruitment error basically means that the equilibrium
equations of the problem have no solution. This can be caused by various
errors in the model including insufficient external boundary conditions
and insufficient muscles inside the body.
They way to debug the model is to first check whether the external
boundary conditions are sufficient. In your case they are, because the
pelvis is fixed in all directions and all rotations. So there is no
doubt that the model as a whole is sufficiently grounded.
Then I started looking for internal problems. In your model this can be
done very easily, because all movement is controlled by joint drivers in
the JointsAndDrivers.any file. Each of the drivers in there contain a
Reaction.Type specification, which for a normal joint is set to {Off},
because the muscles are supposed to carry the load.
However, if the muscles are inufficient, to carry a certain degree of
freedom, switching the corresponding Reaction.Type to {On} will cure the
problem. S0 debugging is a process of trial and error trying to find the
a reaction that, when switched on, will cure the problem.
In your case, the pronation/supination driver on the right forearm seems
to be the problem. It is probably not your fault but rather a case of
the model not having enough muscles in the forearm for the kind of load
you happen to apply to it. The arm model you use stops at the wrist, so
many of the muscles normally crossing the wrist to hands and fingers are
missing, and this is likely why the model cannot carry the load.
The correct fix is to review and improve the mucle configuration in the
forearm. You probaby do not have time for this, so instead you can just
leave the driver’s reaction force on. This will take the load off the
forarm pronators and supinators and allow the algorithm to solve the
problem. In the joints and drivers file, make this change:
//Elbow pronation driver
AnyKinEqInterPolDriver ElbowPronationDriverRight={
AnyKinMeasureOrg &Elbow =…HumanModel.Interface.Right.ElbowPronation;
Data = pi/180*{.JntPos.Right.ElbowPronation };
T=.JntPos.Right.ElbowPronationTime;
Type=Bspline;
BsplineOrder = 8; // This provides smooth accelerations
// Reaction.Type={Off};
Reaction.Type={On};
};
This should make the model run. At least it does so in my version.
Good luck,
John
— In anyscript@yahoogroups.com, Isabel Garcia <isabelgarciamu@…>
wrote:
>
>
>
>
> Hi Soren,
> I’m writing to you from the University of Zaragoza, Spain. Some months
ago, my work mate Raul asked you about an error that anybody returned on
the dynamic analysis. The error was and it continues being the
following:
>
> ERROR(OBJ1) :
D:__AnyBody_LEMUS_Rep6\ARep\Aalborg_LEMUS_Model-LEMUS.Main.any(93) :
‘Study’ : Muscle recruitment analysis failed, simplex solver reports
that solution does not satisfy all constraints.
>
> We are trying to export real movements from a Motion Capture Software
and we are using the FullBodyModel BodyModel.any. Anybody runs the
kinematics analysis and completes it perfectly. The kinematics
constraints are rightly resolved but the inverse dynamics analysis
doesn’t work.
>
> We have applied your previous suggestions but it continues failing in
simple and slowly movements like lifting the right hand and then going
it down. It’s very strange because sometimes when we generate a movement
in Motion Capture, anybody works well, but it can fail in a similar
movement.
>
> I have attached all files we are using in the model, a video with the
movement in the Motion Capture application, the photos generated in
anybody while the kinematics analysis works well, a text file which
contains the error, the files that have been changed from the original
inside Aalborg folder, and finally inside _LEMUS there are the files
generated by MotionCapture. Maybe this helps you to find the mistake.
>
> We have to give in some studies but this error doesn’t allow us to
finish them.
>
> Please, If you have any idea about what’s wrong with that I would be
grateful for your help.
>
> Thank you very much.
>
> Isabel Garcia
>
>
>
> ---------------------------------
> Boardwalk for $500? In 2007? Ha!
> Play Monopoly Here and Now (it’s updated for today’s economy) at
Yahoo! Games.
>
> ---------------------------------
> Don’t pick lemons.
> See all the new 2007 cars at Yahoo! Autos.
>
> [Non-text portions of this message have been removed]
>
[Non-text portions of this message have been removed]
Thank you very much John,
I have changed the muscle configuration in forearm, as you said me, and muscle
recruitment hasn’t failed, but now another error appears:
ERROR(OBJ1): D:_Anybody_LEMUS_Rep6\BRep\Aalborg\Arm3D\Muscle.any(1391):
‘pectoralis_major_clavicular_part_3.SPLine’: Number of allowed iterations for
contact solution has been exceeded in
Main.Model.HumanModel.Right.Shoulder.Arm.Mus.pectoralis_major_clavicular_part_3.
SPLine
What could I do?
Thanks.
Best regards.
Isabel
AnyBody Support <support@anybodytech.com> wrote:
Hello Isabel,
The muscle recruitment error basically means that the equilibrium
equations of the problem have no solution. This can be caused by various
errors in the model including insufficient external boundary conditions
and insufficient muscles inside the body.
They way to debug the model is to first check whether the external
boundary conditions are sufficient. In your case they are, because the
pelvis is fixed in all directions and all rotations. So there is no
doubt that the model as a whole is sufficiently grounded.
Then I started looking for internal problems. In your model this can be
done very easily, because all movement is controlled by joint drivers in
the JointsAndDrivers.any file. Each of the drivers in there contain a
Reaction.Type specification, which for a normal joint is set to {Off},
because the muscles are supposed to carry the load.
However, if the muscles are inufficient, to carry a certain degree of
freedom, switching the corresponding Reaction.Type to {On} will cure the
problem. S0 debugging is a process of trial and error trying to find the
a reaction that, when switched on, will cure the problem.
In your case, the pronation/supination driver on the right forearm seems
to be the problem. It is probably not your fault but rather a case of
the model not having enough muscles in the forearm for the kind of load
you happen to apply to it. The arm model you use stops at the wrist, so
many of the muscles normally crossing the wrist to hands and fingers are
missing, and this is likely why the model cannot carry the load.
The correct fix is to review and improve the mucle configuration in the
forearm. You probaby do not have time for this, so instead you can just
leave the driver’s reaction force on. This will take the load off the
forarm pronators and supinators and allow the algorithm to solve the
problem. In the joints and drivers file, make this change:
//Elbow pronation driver
AnyKinEqInterPolDriver ElbowPronationDriverRight={
AnyKinMeasureOrg &Elbow =…HumanModel.Interface.Right.ElbowPronation;
Data = pi/180*{.JntPos.Right.ElbowPronation };
T=.JntPos.Right.ElbowPronationTime;
Type=Bspline;
BsplineOrder = 8; // This provides smooth accelerations
// Reaction.Type={Off};
Reaction.Type={On};
};
This should make the model run. At least it does so in my version.
Good luck,
John
— In anyscript@yahoogroups.com, Isabel Garcia <isabelgarciamu@…>
wrote:
>
>
>
>
> Hi Soren,
> I’m writing to you from the University of Zaragoza, Spain. Some months
ago, my work mate Raul asked you about an error that anybody returned on
the dynamic analysis. The error was and it continues being the
following:
>
> ERROR(OBJ1) :
D:__AnyBody_LEMUS_Rep6\ARep\Aalborg_LEMUS_Model-LEMUS.Main.any(93) :
‘Study’ : Muscle recruitment analysis failed, simplex solver reports
that solution does not satisfy all constraints.
>
> We are trying to export real movements from a Motion Capture Software
and we are using the FullBodyModel BodyModel.any. Anybody runs the
kinematics analysis and completes it perfectly. The kinematics
constraints are rightly resolved but the inverse dynamics analysis
doesn’t work.
>
> We have applied your previous suggestions but it continues failing in
simple and slowly movements like lifting the right hand and then going
it down. It’s very strange because sometimes when we generate a movement
in Motion Capture, anybody works well, but it can fail in a similar
movement.
>
> I have attached all files we are using in the model, a video with the
movement in the Motion Capture application, the photos generated in
anybody while the kinematics analysis works well, a text file which
contains the error, the files that have been changed from the original
inside Aalborg folder, and finally inside _LEMUS there are the files
generated by MotionCapture. Maybe this helps you to find the mistake.
>
> We have to give in some studies but this error doesn’t allow us to
finish them.
>
> Please, If you have any idea about what’s wrong with that I would be
grateful for your help.
>
> Thank you very much.
>
> Isabel Garcia
>
>
>
> ---------------------------------
> Boardwalk for $500? In 2007? Ha!
> Play Monopoly Here and Now (it’s updated for today’s economy) at
Yahoo! Games.
>
> ---------------------------------
> Don’t pick lemons.
> See all the new 2007 cars at Yahoo! Autos.
>
> [Non-text portions of this message have been removed]
>
[Non-text portions of this message have been removed]
Pinpoint customers who are looking for what you sell.
[Non-text portions of this message have been removed]
Hi Isabel
The error message means the muscle has failed to wrap around objects
in the correct way.
Please double click the error message it will take you to the
Muscle.any file, at the location of the pectoralis muscle.
Normally such errors can be corrected by adding more points on the
muscle, this is done by increasing the Spline.StringMesh property.
This property describes the discretization of the muscle into line
segments used by the wrapping algorithm.
Best regards
Søren, AnyBody Support
— In anyscript@yahoogroups.com, Isabel Garcia <isabelgarciamu@…>
wrote:
>
> Thank you very much John,
> I have changed the muscle configuration in forearm, as you said
me, and muscle recruitment hasn’t failed, but now another error
appears:
>
> ERROR(OBJ1): D:_Anybody_LEMUS_Rep6
\BRep\Aalborg\Arm3D\Muscle.any
(1391): ‘pectoralis_major_clavicular_part_3.SPLine’: Number of
allowed iterations for contact solution has been exceeded in
Main.Model.HumanModel.Right.Shoulder.Arm.Mus.pectoralis_major_clavicu
lar_part_3.SPLine
>
> What could I do?
> Thanks.
> Best regards.
> Isabel
>
> AnyBody Support <support@…> wrote:
>
> Hello Isabel,
>
> The muscle recruitment error basically means that the equilibrium
> equations of the problem have no solution. This can be caused by
various
> errors in the model including insufficient external boundary
conditions
> and insufficient muscles inside the body.
>
> They way to debug the model is to first check whether the external
> boundary conditions are sufficient. In your case they are, because
the
> pelvis is fixed in all directions and all rotations. So there is no
> doubt that the model as a whole is sufficiently grounded.
>
> Then I started looking for internal problems. In your model this
can be
> done very easily, because all movement is controlled by joint
drivers in
> the JointsAndDrivers.any file. Each of the drivers in there
contain a
> Reaction.Type specification, which for a normal joint is set to
{Off},
> because the muscles are supposed to carry the load.
>
> However, if the muscles are inufficient, to carry a certain degree
of
> freedom, switching the corresponding Reaction.Type to {On} will
cure the
> problem. S0 debugging is a process of trial and error trying to
find the
> a reaction that, when switched on, will cure the problem.
>
> In your case, the pronation/supination driver on the right forearm
seems
> to be the problem. It is probably not your fault but rather a case
of
> the model not having enough muscles in the forearm for the kind of
load
> you happen to apply to it. The arm model you use stops at the
wrist, so
> many of the muscles normally crossing the wrist to hands and
fingers are
> missing, and this is likely why the model cannot carry the load.
>
> The correct fix is to review and improve the mucle configuration
in the
> forearm. You probaby do not have time for this, so instead you can
just
> leave the driver’s reaction force on. This will take the load off
the
> forarm pronators and supinators and allow the algorithm to solve
the
> problem. In the joints and drivers file, make this change:
>
> //Elbow pronation driver
>
> AnyKinEqInterPolDriver ElbowPronationDriverRight={
> AnyKinMeasureOrg &Elbow
=…HumanModel.Interface.Right.ElbowPronation;
> Data = pi/180*{.JntPos.Right.ElbowPronation };
> T=.JntPos.Right.ElbowPronationTime;
> Type=Bspline;
> BsplineOrder = 8; // This provides smooth accelerations
>
> // Reaction.Type={Off};
> Reaction.Type={On};
> };
>
> This should make the model run. At least it does so in my version.
>
> Good luck,
> John
>
> — In anyscript@yahoogroups.com, Isabel Garcia <isabelgarciamu@>
> wrote:
> >
> >
> >
> >
> > Hi Soren,
> > I’m writing to you from the University of Zaragoza, Spain. Some
months
> ago, my work mate Raul asked you about an error that anybody
returned on
> the dynamic analysis. The error was and it continues being the
> following:
> >
> > ERROR(OBJ1) :
> D:__AnyBody_LEMUS_Rep6\ARep\Aalborg_LEMUS_Model-LEMUS.Main.any
(93) :
> ‘Study’ : Muscle recruitment analysis failed, simplex solver
reports
> that solution does not satisfy all constraints.
> >
> > We are trying to export real movements from a Motion Capture
Software
> and we are using the FullBodyModel BodyModel.any. Anybody runs the
> kinematics analysis and completes it perfectly. The kinematics
> constraints are rightly resolved but the inverse dynamics analysis
> doesn’t work.
> >
> > We have applied your previous suggestions but it continues
failing in
> simple and slowly movements like lifting the right hand and then
going
> it down. It’s very strange because sometimes when we generate a
movement
> in Motion Capture, anybody works well, but it can fail in a similar
> movement.
> >
> > I have attached all files we are using in the model, a video
with the
> movement in the Motion Capture application, the photos generated in
> anybody while the kinematics analysis works well, a text file which
> contains the error, the files that have been changed from the
original
> inside Aalborg folder, and finally inside _LEMUS there are the
files
> generated by MotionCapture. Maybe this helps you to find the
mistake.
> >
> > We have to give in some studies but this error doesn’t allow us
to
> finish them.
> >
> > Please, If you have any idea about what’s wrong with that I
would be
> grateful for your help.
> >
> > Thank you very much.
> >
> > Isabel Garcia
> >
> >
> >
> > ---------------------------------
> > Boardwalk for $500? In 2007? Ha!
> > Play Monopoly Here and Now (it’s updated for today’s economy) at
> Yahoo! Games.
> >
> > ---------------------------------
> > Don’t pick lemons.
> > See all the new 2007 cars at Yahoo! Autos.
> >
> > [Non-text portions of this message have been removed]
> >
>
> [Non-text portions of this message have been removed]
>
>
>
>
>
>
> ---------------------------------
> Pinpoint customers who are looking for what you sell.
>
> [Non-text portions of this message have been removed]
>
Hi Soren,
I have increased Spline.StringMesh and the error continues appearing,
How much have I to increased it?
Thank you very much.
Isabel
— In anyscript@yahoogroups.com, “AnyBody Support” <support@…> wrote:
>
> Hi Isabel
>
>
> The error message means the muscle has failed to wrap around objects
> in the correct way.
>
> Please double click the error message it will take you to the
> Muscle.any file, at the location of the pectoralis muscle.
>
> Normally such errors can be corrected by adding more points on the
> muscle, this is done by increasing the Spline.StringMesh property.
> This property describes the discretization of the muscle into line
> segments used by the wrapping algorithm.
>
> Best regards
> Søren, AnyBody Support
>
>
>
>
> — In anyscript@yahoogroups.com, Isabel Garcia <isabelgarciamu@>
> wrote:
> >
> > Thank you very much John,
> > I have changed the muscle configuration in forearm, as you said
> me, and muscle recruitment hasn’t failed, but now another error
> appears:
> >
> > ERROR(OBJ1): D:_Anybody_LEMUS_Rep6
> \BRep\Aalborg\Arm3D\Muscle.any
> (1391): ‘pectoralis_major_clavicular_part_3.SPLine’: Number of
> allowed iterations for contact solution has been exceeded in
> Main.Model.HumanModel.Right.Shoulder.Arm.Mus.pectoralis_major_clavicu
> lar_part_3.SPLine
> >
> > What could I do?
> > Thanks.
> > Best regards.
> > Isabel
> >
> > AnyBody Support <support@> wrote:
> >
> > Hello Isabel,
> >
> > The muscle recruitment error basically means that the equilibrium
> > equations of the problem have no solution. This can be caused by
> various
> > errors in the model including insufficient external boundary
> conditions
> > and insufficient muscles inside the body.
> >
> > They way to debug the model is to first check whether the external
> > boundary conditions are sufficient. In your case they are, because
> the
> > pelvis is fixed in all directions and all rotations. So there is no
> > doubt that the model as a whole is sufficiently grounded.
> >
> > Then I started looking for internal problems. In your model this
> can be
> > done very easily, because all movement is controlled by joint
> drivers in
> > the JointsAndDrivers.any file. Each of the drivers in there
> contain a
> > Reaction.Type specification, which for a normal joint is set to
> {Off},
> > because the muscles are supposed to carry the load.
> >
> > However, if the muscles are inufficient, to carry a certain degree
> of
> > freedom, switching the corresponding Reaction.Type to {On} will
> cure the
> > problem. S0 debugging is a process of trial and error trying to
> find the
> > a reaction that, when switched on, will cure the problem.
> >
> > In your case, the pronation/supination driver on the right forearm
> seems
> > to be the problem. It is probably not your fault but rather a case
> of
> > the model not having enough muscles in the forearm for the kind of
> load
> > you happen to apply to it. The arm model you use stops at the
> wrist, so
> > many of the muscles normally crossing the wrist to hands and
> fingers are
> > missing, and this is likely why the model cannot carry the load.
> >
> > The correct fix is to review and improve the mucle configuration
> in the
> > forearm. You probaby do not have time for this, so instead you can
> just
> > leave the driver’s reaction force on. This will take the load off
> the
> > forarm pronators and supinators and allow the algorithm to solve
> the
> > problem. In the joints and drivers file, make this change:
> >
> > //Elbow pronation driver
> >
> > AnyKinEqInterPolDriver ElbowPronationDriverRight={
> > AnyKinMeasureOrg &Elbow
> =…HumanModel.Interface.Right.ElbowPronation;
> > Data = pi/180*{.JntPos.Right.ElbowPronation };
> > T=.JntPos.Right.ElbowPronationTime;
> > Type=Bspline;
> > BsplineOrder = 8; // This provides smooth accelerations
> >
> > // Reaction.Type={Off};
> > Reaction.Type={On};
> > };
> >
> > This should make the model run. At least it does so in my version.
> >
> > Good luck,
> > John
> >
> > — In anyscript@yahoogroups.com, Isabel Garcia <isabelgarciamu@>
> > wrote:
> > >
> > >
> > >
> > >
> > > Hi Soren,
> > > I’m writing to you from the University of Zaragoza, Spain. Some
> months
> > ago, my work mate Raul asked you about an error that anybody
> returned on
> > the dynamic analysis. The error was and it continues being the
> > following:
> > >
> > > ERROR(OBJ1) :
> > D:__AnyBody_LEMUS_Rep6\ARep\Aalborg_LEMUS_Model-LEMUS.Main.any
> (93) :
> > ‘Study’ : Muscle recruitment analysis failed, simplex solver
> reports
> > that solution does not satisfy all constraints.
> > >
> > > We are trying to export real movements from a Motion Capture
> Software
> > and we are using the FullBodyModel BodyModel.any. Anybody runs the
> > kinematics analysis and completes it perfectly. The kinematics
> > constraints are rightly resolved but the inverse dynamics analysis
> > doesn’t work.
> > >
> > > We have applied your previous suggestions but it continues
> failing in
> > simple and slowly movements like lifting the right hand and then
> going
> > it down. It’s very strange because sometimes when we generate a
> movement
> > in Motion Capture, anybody works well, but it can fail in a similar
> > movement.
> > >
> > > I have attached all files we are using in the model, a video
> with the
> > movement in the Motion Capture application, the photos generated in
> > anybody while the kinematics analysis works well, a text file which
> > contains the error, the files that have been changed from the
> original
> > inside Aalborg folder, and finally inside _LEMUS there are the
> files
> > generated by MotionCapture. Maybe this helps you to find the
> mistake.
> > >
> > > We have to give in some studies but this error doesn’t allow us
> to
> > finish them.
> > >
> > > Please, If you have any idea about what’s wrong with that I
> would be
> > grateful for your help.
> > >
> > > Thank you very much.
> > >
> > > Isabel Garcia
> > >
> > >
> > >
> > > ---------------------------------
> > > Boardwalk for $500? In 2007? Ha!
> > > Play Monopoly Here and Now (it’s updated for today’s economy) at
> > Yahoo! Games.
> > >
> > > ---------------------------------
> > > Don’t pick lemons.
> > > See all the new 2007 cars at Yahoo! Autos.
> > >
> > > [Non-text portions of this message have been removed]
> > >
> >
> > [Non-text portions of this message have been removed]
> >
> >
> >
> >
> >
> >
> > ---------------------------------
> > Pinpoint customers who are looking for what you sell.
> >
> > [Non-text portions of this message have been removed]
> >
>
Hi Isabel
Usually increases of 10 or 20 should solve the problem.
At what kind of posture does this error occur at ?
If it is an “extreme” posture for the GH joint, which is the joint
that this muscles spans it could cause problems.
Best regards
Søren
— In anyscript@yahoogroups.com, “isabelgarciamu”
<isabelgarciamu@…> wrote:
>
> Hi Soren,
> I have increased Spline.StringMesh and the error continues
appearing,
> How much have I to increased it?
> Thank you very much.
> Isabel
>
> — In anyscript@yahoogroups.com, “AnyBody Support” <support@>
wrote:
> >
> > Hi Isabel
> >
> >
> > The error message means the muscle has failed to wrap around
objects
> > in the correct way.
> >
> > Please double click the error message it will take you to the
> > Muscle.any file, at the location of the pectoralis muscle.
> >
> > Normally such errors can be corrected by adding more points on
the
> > muscle, this is done by increasing the Spline.StringMesh
property.
> > This property describes the discretization of the muscle into
line
> > segments used by the wrapping algorithm.
> >
> > Best regards
> > Søren, AnyBody Support
> >
> >
> >
> >
> > — In anyscript@yahoogroups.com, Isabel Garcia
<isabelgarciamu@>
> > wrote:
> > >
> > > Thank you very much John,
> > > I have changed the muscle configuration in forearm, as you
said
> > me, and muscle recruitment hasn’t failed, but now another error
> > appears:
> > >
> > > ERROR(OBJ1): D:_Anybody_LEMUS_Rep6
> > \BRep\Aalborg\Arm3D\Muscle.any
> > (1391): ‘pectoralis_major_clavicular_part_3.SPLine’: Number of
> > allowed iterations for contact solution has been exceeded in
> >
Main.Model.HumanModel.Right.Shoulder.Arm.Mus.pectoralis_major_clavicu
> > lar_part_3.SPLine
> > >
> > > What could I do?
> > > Thanks.
> > > Best regards.
> > > Isabel
> > >
> > > AnyBody Support <support@> wrote:
> > >
> > > Hello Isabel,
> > >
> > > The muscle recruitment error basically means that the
equilibrium
> > > equations of the problem have no solution. This can be caused
by
> > various
> > > errors in the model including insufficient external boundary
> > conditions
> > > and insufficient muscles inside the body.
> > >
> > > They way to debug the model is to first check whether the
external
> > > boundary conditions are sufficient. In your case they are,
because
> > the
> > > pelvis is fixed in all directions and all rotations. So there
is no
> > > doubt that the model as a whole is sufficiently grounded.
> > >
> > > Then I started looking for internal problems. In your model
this
> > can be
> > > done very easily, because all movement is controlled by joint
> > drivers in
> > > the JointsAndDrivers.any file. Each of the drivers in there
> > contain a
> > > Reaction.Type specification, which for a normal joint is set
to
> > {Off},
> > > because the muscles are supposed to carry the load.
> > >
> > > However, if the muscles are inufficient, to carry a certain
degree
> > of
> > > freedom, switching the corresponding Reaction.Type to {On}
will
> > cure the
> > > problem. S0 debugging is a process of trial and error trying
to
> > find the
> > > a reaction that, when switched on, will cure the problem.
> > >
> > > In your case, the pronation/supination driver on the right
forearm
> > seems
> > > to be the problem. It is probably not your fault but rather a
case
> > of
> > > the model not having enough muscles in the forearm for the
kind of
> > load
> > > you happen to apply to it. The arm model you use stops at the
> > wrist, so
> > > many of the muscles normally crossing the wrist to hands and
> > fingers are
> > > missing, and this is likely why the model cannot carry the
load.
> > >
> > > The correct fix is to review and improve the mucle
configuration
> > in the
> > > forearm. You probaby do not have time for this, so instead you
can
> > just
> > > leave the driver’s reaction force on. This will take the load
off
> > the
> > > forarm pronators and supinators and allow the algorithm to
solve
> > the
> > > problem. In the joints and drivers file, make this change:
> > >
> > > //Elbow pronation driver
> > >
> > > AnyKinEqInterPolDriver ElbowPronationDriverRight={
> > > AnyKinMeasureOrg &Elbow
> > =…HumanModel.Interface.Right.ElbowPronation;
> > > Data = pi/180*{.JntPos.Right.ElbowPronation };
> > > T=.JntPos.Right.ElbowPronationTime;
> > > Type=Bspline;
> > > BsplineOrder = 8; // This provides smooth accelerations
> > >
> > > // Reaction.Type={Off};
> > > Reaction.Type={On};
> > > };
> > >
> > > This should make the model run. At least it does so in my
version.
> > >
> > > Good luck,
> > > John
> > >
> > > — In anyscript@yahoogroups.com, Isabel Garcia
<isabelgarciamu@>
> > > wrote:
> > > >
> > > >
> > > >
> > > >
> > > > Hi Soren,
> > > > I’m writing to you from the University of Zaragoza, Spain.
Some
> > months
> > > ago, my work mate Raul asked you about an error that anybody
> > returned on
> > > the dynamic analysis. The error was and it continues being the
> > > following:
> > > >
> > > > ERROR(OBJ1) :
> > > D:__AnyBody_LEMUS_Rep6\ARep\Aalborg_LEMUS_Model-
LEMUS.Main.any
> > (93) :
> > > ‘Study’ : Muscle recruitment analysis failed, simplex solver
> > reports
> > > that solution does not satisfy all constraints.
> > > >
> > > > We are trying to export real movements from a Motion Capture
> > Software
> > > and we are using the FullBodyModel BodyModel.any. Anybody runs
the
> > > kinematics analysis and completes it perfectly. The kinematics
> > > constraints are rightly resolved but the inverse dynamics
analysis
> > > doesn’t work.
> > > >
> > > > We have applied your previous suggestions but it continues
> > failing in
> > > simple and slowly movements like lifting the right hand and
then
> > going
> > > it down. It’s very strange because sometimes when we generate
a
> > movement
> > > in Motion Capture, anybody works well, but it can fail in a
similar
> > > movement.
> > > >
> > > > I have attached all files we are using in the model, a video
> > with the
> > > movement in the Motion Capture application, the photos
generated in
> > > anybody while the kinematics analysis works well, a text file
which
> > > contains the error, the files that have been changed from the
> > original
> > > inside Aalborg folder, and finally inside _LEMUS there are the
> > files
> > > generated by MotionCapture. Maybe this helps you to find the
> > mistake.
> > > >
> > > > We have to give in some studies but this error doesn’t allow
us
> > to
> > > finish them.
> > > >
> > > > Please, If you have any idea about what’s wrong with that I
> > would be
> > > grateful for your help.
> > > >
> > > > Thank you very much.
> > > >
> > > > Isabel Garcia
> > > >
> > > >
> > > >
> > > > ---------------------------------
> > > > Boardwalk for $500? In 2007? Ha!
> > > > Play Monopoly Here and Now (it’s updated for today’s
economy) at
> > > Yahoo! Games.
> > > >
> > > > ---------------------------------
> > > > Don’t pick lemons.
> > > > See all the new 2007 cars at Yahoo! Autos.
> > > >
> > > > [Non-text portions of this message have been removed]
> > > >
> > >
> > > [Non-text portions of this message have been removed]
> > >
> > >
> > >
> > >
> > >
> > >
> > > ---------------------------------
> > > Pinpoint customers who are looking for what you sell.
> > >
> > > [Non-text portions of this message have been removed]
> > >
> >
>