Would you recommend using global or local coordinates to drive a model
with markers? For example, in the gait model, was that driven by
markers as they move through a global environment or locally in
comparison with each other?
Which would you recommend?
Thanks,
Sarah
Hi Sarah
It is a good question
The gait model is driven using the global coordinate system for
measuring between the human model markers and the free floating
measured markers, except for the Metatarsal marker which is using
the
local. This works well for the gait model since it does not turn
around wrt. the global coordinate system.
In general i think the local formulation is the best so the local
coordinate systems of the bony segments should be used. We have done
some experiments on a model where we did that and it worked out
well.
The advantage of this setup is that when the person turns around
wrt.
to the global coordinate system this will not affect how the model
is
driven model. If it uses the global one it will have a big impact.
So introduce a Ref=0 in the AnyKinLinear’s you are using for driving
the model. Then it will use the first of the reference frames in the
measure as the reference coordinate system.
Best regards
AnyBody Support
— In anyscript@yahoogroups.com, “Sarah R. Sullivan”
<sarsulli@e…>
wrote:
>
> Would you recommend using global or local coordinates to drive a
model
> with markers? For example, in the gait model, was that driven by
> markers as they move through a global environment or locally in
> comparison with each other?
>
> Which would you recommend?
> Thanks,
> Sarah
>
Hi,
Just to make sure I get this right. Let’s say a set of markers that
are reference to a GlobalMarkerRef and a model that is referenced to a
GlobalRef. If, for the linear measure:
I use a local referencing system with the model node listed first,
Anybody will convert the marker’s coordinate from the GlobalMarkerRef
to the NodeRef and will compare the position of the marker to the
origin of the NodeRef.
I use a global referencing system, Anybody will convert both the
node position and the marker position to the GlobalRef and compare the
distance between the two.
Is that right?
Jean-Olivier
— In anyscript@yahoogroups.com, “AnyBody Support” <support@a…> wrote:
>
> Hi Sarah
>
> It is a good question
>
> The gait model is driven using the global coordinate system for
> measuring between the human model markers and the free floating
> measured markers, except for the Metatarsal marker which is using
> the
> local. This works well for the gait model since it does not turn
> around wrt. the global coordinate system.
>
> In general i think the local formulation is the best so the local
> coordinate systems of the bony segments should be used. We have done
> some experiments on a model where we did that and it worked out
> well.
> The advantage of this setup is that when the person turns around
> wrt.
> to the global coordinate system this will not affect how the model
> is
> driven model. If it uses the global one it will have a big impact.
>
> So introduce a Ref=0 in the AnyKinLinear’s you are using for driving
> the model. Then it will use the first of the reference frames in the
> measure as the reference coordinate system.
>
> Best regards
> AnyBody Support
>
> — In anyscript@yahoogroups.com, “Sarah R. Sullivan”
> <sarsulli@e…>
> wrote:
> >
> > Would you recommend using global or local coordinates to drive a
> model
> > with markers? For example, in the gait model, was that driven by
> > markers as they move through a global environment or locally in
> > comparison with each other?
> >
> > Which would you recommend?
> > Thanks,
> > Sarah
> >
>
Hi Oliver
I will try to explain
If you write
AnyKinLinear marker={
AnyRefFrame &marker = …
AnyRefFrame &markeronhuman = …
};
this measure will measure the difference in x,y,z between the two
markers using the global reference system,
AnyKinLinear marker={
AnyRefFrame &marker = …
AnyRefFrame &markeronhuman = …
Ref=0;
};
this measure will measure the difference in x,y,z between the two
markers using the marker reference system. In the gait model it will
have same orientation as the global ref so this will measure the
same way as the first example
AnyKinLinear marker={
AnyRefFrame &humanmarker = …
AnyRefFrame &marker = …
Ref=0;
};
this measure will measure the difference in x,y,z between the two
markers using the humanmarker reference system. This is very
different since this coordinate system changes orientation when the
leg is moving.
For more details please look up AnyKinLinear in the reference manual
http://www.anybodytech.com/fileadmin/doc/AnyBodyRefManual.pdf
Please do not hesitate to ask again if you have further questions.
Best regards
AnyBody Support
— In anyscript@yahoogroups.com, “jeanolivierracine” <jean-
olivier.racine.1@e…> wrote:
>
> Hi,
> Just to make sure I get this right. Let’s say a set of markers that
> are reference to a GlobalMarkerRef and a model that is referenced
to a
> GlobalRef. If, for the linear measure:
>
> - I use a local referencing system with the model node listed
first,
> Anybody will convert the marker’s coordinate from the
GlobalMarkerRef
> to the NodeRef and will compare the position of the marker to the
> origin of the NodeRef.
>
> - I use a global referencing system, Anybody will convert both the
> node position and the marker position to the GlobalRef and compare
the
> distance between the two.
>
> Is that right?
>
> Jean-Olivier
>
>
> — In anyscript@yahoogroups.com, “AnyBody Support” <support@a…>
wrote:
> >
> > Hi Sarah
> >
> > It is a good question
> >
> > The gait model is driven using the global coordinate system for
> > measuring between the human model markers and the free floating
> > measured markers, except for the Metatarsal marker which is
using
> > the
> > local. This works well for the gait model since it does not turn
> > around wrt. the global coordinate system.
> >
> > In general i think the local formulation is the best so the
local
> > coordinate systems of the bony segments should be used. We have
done
> > some experiments on a model where we did that and it worked out
> > well.
> > The advantage of this setup is that when the person turns around
> > wrt.
> > to the global coordinate system this will not affect how the
model
> > is
> > driven model. If it uses the global one it will have a big
impact.
> >
> > So introduce a Ref=0 in the AnyKinLinear’s you are using for
driving
> > the model. Then it will use the first of the reference frames in
the
> > measure as the reference coordinate system.
> >
> > Best regards
> > AnyBody Support
> >
> > — In anyscript@yahoogroups.com, “Sarah R. Sullivan”
> > <sarsulli@e…>
> > wrote:
> > >
> > > Would you recommend using global or local coordinates to drive
a
> > model
> > > with markers? For example, in the gait model, was that driven
by
> > > markers as they move through a global environment or locally in
> > > comparison with each other?
> > >
> > > Which would you recommend?
> > > Thanks,
> > > Sarah
> > >
> >
>
Great, sounds good, I get the idea.
But it brings up another question:
Suppose we go back to the second example, but use it in a driver context:
AnyKinEqSimpleDriver markerdriver = {
AnyKinLinear marker={
AnyRefFrame &humanmarker = …
AnyRefFrame &marker = …
Ref=0;
};
MeasureOrganizer = {0, 2};
…
Now, suppose humanmarker has a different orientation than the
GlobalRef. When I use the organizer and specified {0, 2}, was I
refering to the GlobalRef or the humanmarker reference system? In
other words, is it the X and Z coordinates according to the
humanmarker or GlobalRef that I will be driving? The scope suggests
that it will be GlobalRef, but again…
— In anyscript@yahoogroups.com, “AnyBody Support” <support@a…> wrote:
>
> Hi Oliver
>
> I will try to explain
>
> If you write
>
> AnyKinLinear marker={
> AnyRefFrame &marker = …
> AnyRefFrame &markeronhuman = …
> };
>
> this measure will measure the difference in x,y,z between the two
> markers using the global reference system,
>
> AnyKinLinear marker={
> AnyRefFrame &marker = …
> AnyRefFrame &markeronhuman = …
> Ref=0;
> };
>
>
> this measure will measure the difference in x,y,z between the two
> markers using the marker reference system. In the gait model it will
> have same orientation as the global ref so this will measure the
> same way as the first example
>
> AnyKinLinear marker={
> AnyRefFrame &humanmarker = …
> AnyRefFrame &marker = …
> Ref=0;
> };
>
>
> this measure will measure the difference in x,y,z between the two
> markers using the humanmarker reference system. This is very
> different since this coordinate system changes orientation when the
> leg is moving.
>
> For more details please look up AnyKinLinear in the reference manual
>
> http://www.anybodytech.com/fileadmin/doc/AnyBodyRefManual.pdf
>
> Please do not hesitate to ask again if you have further questions.
>
> Best regards
> AnyBody Support
>
> — In anyscript@yahoogroups.com, “jeanolivierracine” <jean-
> olivier.racine.1@e…> wrote:
> >
> > Hi,
> > Just to make sure I get this right. Let’s say a set of markers that
> > are reference to a GlobalMarkerRef and a model that is referenced
> to a
> > GlobalRef. If, for the linear measure:
> >
> > - I use a local referencing system with the model node listed
> first,
> > Anybody will convert the marker’s coordinate from the
> GlobalMarkerRef
> > to the NodeRef and will compare the position of the marker to the
> > origin of the NodeRef.
> >
> > - I use a global referencing system, Anybody will convert both the
> > node position and the marker position to the GlobalRef and compare
> the
> > distance between the two.
> >
> > Is that right?
> >
> > Jean-Olivier
> >
> >
> > — In anyscript@yahoogroups.com, “AnyBody Support” <support@a…>
> wrote:
> > >
> > > Hi Sarah
> > >
> > > It is a good question
> > >
> > > The gait model is driven using the global coordinate system for
> > > measuring between the human model markers and the free floating
> > > measured markers, except for the Metatarsal marker which is
> using
> > > the
> > > local. This works well for the gait model since it does not turn
> > > around wrt. the global coordinate system.
> > >
> > > In general i think the local formulation is the best so the
> local
> > > coordinate systems of the bony segments should be used. We have
> done
> > > some experiments on a model where we did that and it worked out
> > > well.
> > > The advantage of this setup is that when the person turns around
> > > wrt.
> > > to the global coordinate system this will not affect how the
> model
> > > is
> > > driven model. If it uses the global one it will have a big
> impact.
> > >
> > > So introduce a Ref=0 in the AnyKinLinear’s you are using for
> driving
> > > the model. Then it will use the first of the reference frames in
> the
> > > measure as the reference coordinate system.
> > >
> > > Best regards
> > > AnyBody Support
> > >
> > > — In anyscript@yahoogroups.com, “Sarah R. Sullivan”
> > > <sarsulli@e…>
> > > wrote:
> > > >
> > > > Would you recommend using global or local coordinates to drive
> a
> > > model
> > > > with markers? For example, in the gait model, was that driven
> by
> > > > markers as they move through a global environment or locally in
> > > > comparison with each other?
> > > >
> > > > Which would you recommend?
> > > > Thanks,
> > > > Sarah
> > > >
> > >
> >
>
Hi Oliver
It will drive x difference and z difference between the to nodes,
using the coordinate system in the humanmarker, not the global
cooridnate system. So the the coordiantesystem will be moving with
the human bone and change orientation when the person moves.
This setup is good because you will always drive the model in the
same way even if he rotates in the global reference system. This is
not the case when the globalref is being used.
As an example take the metatarsal maker on the foot if you drive
this is x direction of the foot coordinate system, which points
upwards for a stading model. This dof will be good for driving the
ankle flexion no matter how the model is orientatet in the
globalref.
Best regards
AnyBody Support
— In anyscript@yahoogroups.com, “jeanolivierracine” <jean-
olivier.racine.1@e…> wrote:
>
> Great, sounds good, I get the idea.
>
> But it brings up another question:
>
> Suppose we go back to the second example, but use it in a driver
context:
>
> AnyKinEqSimpleDriver markerdriver = {
> AnyKinLinear marker={
> AnyRefFrame &humanmarker = …
> AnyRefFrame &marker = …
> Ref=0;
> };
> MeasureOrganizer = {0, 2};
> …
>
> Now, suppose humanmarker has a different orientation than the
> GlobalRef. When I use the organizer and specified {0, 2}, was I
> refering to the GlobalRef or the humanmarker reference system? In
> other words, is it the X and Z coordinates according to the
> humanmarker or GlobalRef that I will be driving? The scope suggests
> that it will be GlobalRef, but again…
>
> — In anyscript@yahoogroups.com, “AnyBody Support” <support@a…>
wrote:
> >
> > Hi Oliver
> >
> > I will try to explain
> >
> > If you write
> >
> > AnyKinLinear marker={
> > AnyRefFrame &marker = …
> > AnyRefFrame &markeronhuman = …
> > };
> >
> > this measure will measure the difference in x,y,z between the
two
> > markers using the global reference system,
> >
> > AnyKinLinear marker={
> > AnyRefFrame &marker = …
> > AnyRefFrame &markeronhuman = …
> > Ref=0;
> > };
> >
> >
> > this measure will measure the difference in x,y,z between the
two
> > markers using the marker reference system. In the gait model it
will
> > have same orientation as the global ref so this will measure the
> > same way as the first example
> >
> > AnyKinLinear marker={
> > AnyRefFrame &humanmarker = …
> > AnyRefFrame &marker = …
> > Ref=0;
> > };
> >
> >
> > this measure will measure the difference in x,y,z between the
two
> > markers using the humanmarker reference system. This is very
> > different since this coordinate system changes orientation when
the
> > leg is moving.
> >
> > For more details please look up AnyKinLinear in the reference
manual
> >
> > http://www.anybodytech.com/fileadmin/doc/AnyBodyRefManual.pdf
> >
> > Please do not hesitate to ask again if you have further
questions.
> >
> > Best regards
> > AnyBody Support
> >
> > — In anyscript@yahoogroups.com, “jeanolivierracine” <jean-
> > olivier.racine.1@e…> wrote:
> > >
> > > Hi,
> > > Just to make sure I get this right. Let’s say a set of markers
that
> > > are reference to a GlobalMarkerRef and a model that is
referenced
> > to a
> > > GlobalRef. If, for the linear measure:
> > >
> > > - I use a local referencing system with the model node listed
> > first,
> > > Anybody will convert the marker’s coordinate from the
> > GlobalMarkerRef
> > > to the NodeRef and will compare the position of the marker to
the
> > > origin of the NodeRef.
> > >
> > > - I use a global referencing system, Anybody will convert both
the
> > > node position and the marker position to the GlobalRef and
compare
> > the
> > > distance between the two.
> > >
> > > Is that right?
> > >
> > > Jean-Olivier
> > >
> > >
> > > — In anyscript@yahoogroups.com, “AnyBody Support”
<support@a…>
> > wrote:
> > > >
> > > > Hi Sarah
> > > >
> > > > It is a good question
> > > >
> > > > The gait model is driven using the global coordinate system
for
> > > > measuring between the human model markers and the free
floating
> > > > measured markers, except for the Metatarsal marker which is
> > using
> > > > the
> > > > local. This works well for the gait model since it does not
turn
> > > > around wrt. the global coordinate system.
> > > >
> > > > In general i think the local formulation is the best so the
> > local
> > > > coordinate systems of the bony segments should be used. We
have
> > done
> > > > some experiments on a model where we did that and it worked
out
> > > > well.
> > > > The advantage of this setup is that when the person turns
around
> > > > wrt.
> > > > to the global coordinate system this will not affect how the
> > model
> > > > is
> > > > driven model. If it uses the global one it will have a big
> > impact.
> > > >
> > > > So introduce a Ref=0 in the AnyKinLinear’s you are using for
> > driving
> > > > the model. Then it will use the first of the reference
frames in
> > the
> > > > measure as the reference coordinate system.
> > > >
> > > > Best regards
> > > > AnyBody Support
> > > >
> > > > — In anyscript@yahoogroups.com, “Sarah R. Sullivan”
> > > > <sarsulli@e…>
> > > > wrote:
> > > > >
> > > > > Would you recommend using global or local coordinates to
drive
> > a
> > > > model
> > > > > with markers? For example, in the gait model, was that
driven
> > by
> > > > > markers as they move through a global environment or
locally in
> > > > > comparison with each other?
> > > > >
> > > > > Which would you recommend?
> > > > > Thanks,
> > > > > Sarah
> > > > >
> > > >
> > >
> >
>