Driven by markers

Good afternoon,

I am trying to drive the StandingModel (full body) using markers. As a
guide, I use the Gait3D model and the Gait webcast presentation. Here is my
question:

In the presentation, the green arrows correspond to the last constraint, the
SacrumZ. The equation used is RAsisZ + LAsisZ = SacrumZ. According to the
figure in p.10 of the presentation, the Z axis is vertical. Therefore, it
would mean that the height of the sacrum needs to be twice the average
height of the Asis. What am I not getting right?

Thanks,
Jean-Olivier
LIO - CHUM

Hi Jean-Oliver

RAsisZ is a measure in the z direction betwen the right Asis marker on the model
and the free floating makrer

LAsisZ is a measure in the z direction betwen the Left Asis marker on the model
and the free floating makrer

SacrumZ is a measure in the z direction betwen the Sacrum marker on the model
and the free floating makrer

So the equation RAsisZ + LAsisZ = SacrumZ means that distance measured in the z
direction between the human marker and the free floating asis markers needs to
add up to be exact the error on the sacrum marker measured in z. So what we are
trying to do is to distribute the error between the three markers. It do not
mean that " the height of the sacrum needs to be twice the average height of
the Asis" it means that distances measured in the z direction between the model
markers and the free floating markers needs to be that be given according to the
equation.

It sounds as an interesting model you are building here are some advices:

Try to keep having a running model by using joint drivers on most joints and
then gradually one by one replace those by marker drivers.

Switch off the muscles when you play around with the kinematics, this can be
done from the main file.

Always keep a exact count on the number of drivers you add to the model and
remove, this number should never change, for the same body model.

Please do not hesitate to write agian if you have further questions.

Best regards

AnyBody Support

— In anyscript@yahoogroups.com, “Jean-Olivier Racine”
<jean-olivier.racine.1@e…> wrote:
>
> Good afternoon,
>
> I am trying to drive the StandingModel (full body) using markers. As a
> guide, I use the Gait3D model and the Gait webcast presentation. Here is my
> question:
>
> In the presentation, the green arrows correspond to the last constraint, the
> SacrumZ. The equation used is RAsisZ + LAsisZ = SacrumZ. According to the
> figure in p.10 of the presentation, the Z axis is vertical. Therefore, it
> would mean that the height of the sacrum needs to be twice the average
> height of the Asis. What am I not getting right?
>
> Thanks,
> Jean-Olivier
> LIO - CHUM
>

Hi again,

I do have more questions.

As you suggested, I tried to add the markers one by one.
Unfortunately, it looks like it can’t be done this way because the
“fullbodymodel”'s position is driven by it’s CoM. This is incompatible
with my marker’s positions.

So I tried to constraint it as follow:

  • Right and left heel: Y (2)
  • Right and left maleolus: X, Y, Z (6)
  • Right hip add/abd = Left hip add/abd: X (1)
  • Right ASIS: Z (1)
    TOTAL: 10

This replaces these drivers:

  • Right and left toe: Y (2)
  • Right and left heel: Y (2)
  • Right and left ankle: X (2)
  • CoM on global ref: X, Z (2)
  • Right and left hip add/abd: X (2)
    TOTAL: 10

NOTE: All other drivers are unchanged.

The model compiles and loads without over or under determinacy of the
kinematics constraints. But when I try to set the initial conditions,
I get the “constraint above error tolerance” errors.

My questions:

  1. Can we assume that the first constraint that is named in the list
    of constraints that are above tolerance is the source of the problem?
  2. What’s wrong with my logic?

NOTE: I uploaded my model source to the file section, as two (2) RAR
files called ARep-LIO.RAR and BRep-LIO.RAR under “Jean-Olivier’s Project”.

Thank you very much!
Jean-Olivier

— In anyscript@yahoogroups.com, “AnyBody Support” <support@a…> wrote:
>
>
> Hi Jean-Oliver
>
> RAsisZ is a measure in the z direction betwen the right Asis marker
on the model and the free floating makrer
>
> LAsisZ is a measure in the z direction betwen the Left Asis marker
on the model and the free floating makrer
>
> SacrumZ is a measure in the z direction betwen the Sacrum marker on
the model and the free floating makrer
>
> So the equation RAsisZ + LAsisZ = SacrumZ means that distance
measured in the z direction between the human marker and the free
floating asis markers needs to add up to be exact the error on the
sacrum marker measured in z. So what we are trying to do is to
distribute the error between the three markers. It do not mean that
" the height of the sacrum needs to be twice the average height of the
Asis" it means that distances measured in the z direction between the
model markers and the free floating markers needs to be that be given
according to the equation.
>
> It sounds as an interesting model you are building here are some
advices:
>
> Try to keep having a running model by using joint drivers on most
joints and then gradually one by one replace those by marker drivers.
>
> Switch off the muscles when you play around with the kinematics,
this can be done from the main file.
>
> Always keep a exact count on the number of drivers you add to the
model and remove, this number should never change, for the same body
model.
>
> Please do not hesitate to write agian if you have further questions.
>
> Best regards
>
> AnyBody Support
>
>
>
>
>
>
>
>
>
> — In anyscript@yahoogroups.com, “Jean-Olivier Racine”
<jean-olivier.racine.1@e…> wrote:
> >
> > Good afternoon,
> >
> > I am trying to drive the StandingModel (full body) using markers. As a
> > guide, I use the Gait3D model and the Gait webcast presentation.
Here is my
> > question:
> >
> > In the presentation, the green arrows correspond to the last
constraint, the
> > SacrumZ. The equation used is RAsisZ + LAsisZ = SacrumZ. According
to the
> > figure in p.10 of the presentation, the Z axis is vertical.
Therefore, it
> > would mean that the height of the sacrum needs to be twice the average
> > height of the Asis. What am I not getting right?
> >
> > Thanks,
> > Jean-Olivier
> > LIO - CHUM
> >
>

Hi Oliver

I have been looking at your model, and uploaded a new
JointAndDrivers.any file to your directory, please search for
aybodysup to see the changes.

There are several problems, i have removed joint drivers on the
legs, the legs where driven by both markers and joint drivers. I do
not think it was intentional, but i might be wrong.

I think you should still try to make it in smaller steps.

Include the file

#include “…\BRep\Aalborg\BodyModels\TwoLegs\BodyModel_Mus3E.a
ny”

instead of the full body model and make sure that this is driven
correctly before you move on to the fullbody.

Make sure you know the orientation of the local coordinate systems
if you use them. It is a good idea to make a small sketch of the
model and how it is driven. But i would start by driving it like we
do on the gait model if possible.

Using the legs only you should use 18 divers, you can see how the
gait model is driven in this presentation

http://www.anybodytech.com/fileadmin/examples/GaitWebcast.pdf
(you need to log in to see it)

If you make a stetch on the drivers you use i would be happy to see,
and comment it.

I hope this moves you in the right direction, even though it does
not solves all the problems.

About the error message: in general look for the largest number,
this will normally give you a good indication of the root fo the
problem.

Best regards
AnyBody support

— In anyscript@yahoogroups.com, “jeanolivierracine” <jean-
olivier.racine.1@e…> wrote:
>
> Hi again,
>
> I do have more questions.
>
> As you suggested, I tried to add the markers one by one.
> Unfortunately, it looks like it can’t be done this way because the
> “fullbodymodel”'s position is driven by it’s CoM. This is
incompatible
> with my marker’s positions.
>
> So I tried to constraint it as follow:
> - Right and left heel: Y (2)
> - Right and left maleolus: X, Y, Z (6)
> - Right hip add/abd = Left hip add/abd: X (1)
> - Right ASIS: Z (1)
> TOTAL: 10
>
> This replaces these drivers:
> - Right and left toe: Y (2)
> - Right and left heel: Y (2)
> - Right and left ankle: X (2)
> - CoM on global ref: X, Z (2)
> - Right and left hip add/abd: X (2)
> TOTAL: 10
>
> NOTE: All other drivers are unchanged.
>
> The model compiles and loads without over or under determinacy of
the
> kinematics constraints. But when I try to set the initial
conditions,
> I get the “constraint above error tolerance” errors.
>
> My questions:
> 1. Can we assume that the first constraint that is named in the
list
> of constraints that are above tolerance is the source of the
problem?
> 2. What’s wrong with my logic?
>
> NOTE: I uploaded my model source to the file section, as two (2)
RAR
> files called ARep-LIO.RAR and BRep-LIO.RAR under “Jean-Olivier’s
Project”.
>
> Thank you very much!
> Jean-Olivier
>
> — In anyscript@yahoogroups.com, “AnyBody Support” <support@a…>
wrote:
> >
> >
> > Hi Jean-Oliver
> >
> > RAsisZ is a measure in the z direction betwen the right Asis
marker
> on the model and the free floating makrer
> >
> > LAsisZ is a measure in the z direction betwen the Left Asis
marker
> on the model and the free floating makrer
> >
> > SacrumZ is a measure in the z direction betwen the Sacrum marker
on
> the model and the free floating makrer
> >
> > So the equation RAsisZ + LAsisZ = SacrumZ means that distance
> measured in the z direction between the human marker and the free
> floating asis markers needs to add up to be exact the error on the
> sacrum marker measured in z. So what we are trying to do is to
> distribute the error between the three markers. It do not mean
that
> " the height of the sacrum needs to be twice the average height of
the
> Asis" it means that distances measured in the z direction between
the
> model markers and the free floating markers needs to be that be
given
> according to the equation.
> >
> > It sounds as an interesting model you are building here are some
> advices:
> >
> > Try to keep having a running model by using joint drivers on most
> joints and then gradually one by one replace those by marker
drivers.
> >
> > Switch off the muscles when you play around with the kinematics,
> this can be done from the main file.
> >
> > Always keep a exact count on the number of drivers you add to the
> model and remove, this number should never change, for the same
body
> model.
> >
> > Please do not hesitate to write agian if you have further
questions.
> >
> > Best regards
> >
> > AnyBody Support
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > — In anyscript@yahoogroups.com, “Jean-Olivier Racine”
> <jean-olivier.racine.1@e…> wrote:
> > >
> > > Good afternoon,
> > >
> > > I am trying to drive the StandingModel (full body) using
markers. As a
> > > guide, I use the Gait3D model and the Gait webcast
presentation.
> Here is my
> > > question:
> > >
> > > In the presentation, the green arrows correspond to the last
> constraint, the
> > > SacrumZ. The equation used is RAsisZ + LAsisZ = SacrumZ.
According
> to the
> > > figure in p.10 of the presentation, the Z axis is vertical.
> Therefore, it
> > > would mean that the height of the sacrum needs to be twice the
average
> > > height of the Asis. What am I not getting right?
> > >
> > > Thanks,
> > > Jean-Olivier
> > > LIO - CHUM
> > >
> >
>