Hi everyone,
I’ve been collecting gait data for my project and trying to run
kinematic analysis. i have been setting up my joints and drivers
folder to constrain/drive the model. I am finding it hard to visualise
how to drive/constrain each degree of freedom. As a result i am using
trial and error in order to try and run the model. Is there any
function that allows me to check what plane a DOF needs to be driven in?
many thanks
Peter
Hi Peter
My experience is that using trial and error mostly creates errors:D I
think it is important that the model always runs, so start letting
everything be “not driven” and then you add/remove a DOF once at a
time. Depending on how you drive the model it is always good to be
able to see the axis of the coordinate system in the reference frame
you drive it in. It is always a good idea to write down in what DOF
that you want to drive the different segments, in that way you can
sort of plan your programming:D
What model did you start from? the gait3d model?
Hope you get it right:D else please post again.
Best regards
Christian, AnyBody Support
— In anyscript@yahoogroups.com, “peter.worsley” <peter.worsley@…>
wrote:
>
> Hi everyone,
>
> I’ve been collecting gait data for my project and trying to run
> kinematic analysis. i have been setting up my joints and drivers
> folder to constrain/drive the model. I am finding it hard to visualise
> how to drive/constrain each degree of freedom. As a result i am using
> trial and error in order to try and run the model. Is there any
> function that allows me to check what plane a DOF needs to be driven in?
>
> many thanks
>
> Peter
>
Hi Christian,
I tried to take your advice and I put in AnyStdJoints to stop my DOF
driver error for kinematic analysis. But I am still getting a
indeterminant system. I have based my model on the Aalborg gait 3D,
but I notice that you have added a sacral marker set in the model (i
am using PSIS). I think the problem is lying in the pelvis, I have
contrained 160 out of 161 contraints and tried to make a sacral marker
but with no luck (its coming up with unsovlable constraint, when i
run the model).
Your help would be greatly appreciated as always
many thanks
Peter
— In anyscript@yahoogroups.com, “AnyBody Support” <support@…> wrote:
>
> Hi Peter
>
> My experience is that using trial and error mostly creates errors:D I
> think it is important that the model always runs, so start letting
> everything be “not driven” and then you add/remove a DOF once at a
> time. Depending on how you drive the model it is always good to be
> able to see the axis of the coordinate system in the reference frame
> you drive it in. It is always a good idea to write down in what DOF
> that you want to drive the different segments, in that way you can
> sort of plan your programming:D
> What model did you start from? the gait3d model?
>
> Hope you get it right:D else please post again.
>
> Best regards
> Christian, AnyBody Support
>
> — In anyscript@yahoogroups.com, “peter.worsley” <peter.worsley@>
> wrote:
> >
> > Hi everyone,
> >
> > I’ve been collecting gait data for my project and trying to run
> > kinematic analysis. i have been setting up my joints and drivers
> > folder to constrain/drive the model. I am finding it hard to visualise
> > how to drive/constrain each degree of freedom. As a result i am using
> > trial and error in order to try and run the model. Is there any
> > function that allows me to check what plane a DOF needs to be
driven in?
> >
> > many thanks
> >
> > Peter
> >
>
Hi Peter
If you think that the problem is related to the pelvis I suggest that you
insert joint drivers for the hip, knee and ankle joint and only keep the
pelvis drivers, if it still does not run and the pelvis has drivers for six
dof then the problem is related to driving the pelvis, and you have isolated
the problem.
If this does not work then in line with what Christian wrote i would suggest
the following step procedure:
1: Take the freeposture model replace the bodymodel with the twolegs model
remove the unwanted drivers. This should give you a model where all joints
are driven.
2: Scale the model so the size is fitting place markers on the right
location of the bones.
3: By making changes to the mannequin.any file, position the model so that
it fit in the best possible way with the marker locations of the recorded
markers. This will give you a good initial position guess which is important
for marker driven model like this one. 3: Now remove some of the drivers
from the pelvis and replace them gradually by marker drivers, for each step
you do make sure the kinematics are running all the way through.
The two leg gait model has the following dof’s: 6 rigidbody dof of pelvis, 3
hip, 1 knee, 2 ankle in total for both legs this gives 18, so you need
exactly 18 drivers, and they must be able to drive these dofs. When you
select which dof to driver for a specific marker it is usually a good idea
to select a direction which is normal to the long axis of the bone the
marker is attached too. Here it is important to understand the concept of
the AnyKinLinear measure and the difference of using a Ref=0 or Ref=-1,
please refer to the manual if needed. You will get the best solution is you
use Ref=0, because this makes the model independent of the orientation of
the subject. Sometimes I do a small stick figure sketch which shows the
drivers as arrows pointing in the right directions according to the segment
coordinate system.
4: continue to remove the joint drivers replacing them by the marker
drivers, one by one until they are all replaced.
I hope this helps you move on.
Best regards
Søren, AnyBody Support
From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 12 February 2008 13:21
To: anyscript@yahoogroups.com
Subject: [AnyScript] Re: kinematic indeterminate systems
Hi Christian,
I tried to take your advice and I put in AnyStdJoints to stop my DOF
driver error for kinematic analysis. But I am still getting a
indeterminant system. I have based my model on the Aalborg gait 3D,
but I notice that you have added a sacral marker set in the model (i
am using PSIS). I think the problem is lying in the pelvis, I have
contrained 160 out of 161 contraints and tried to make a sacral marker
but with no luck (its coming up with unsovlable constraint, when i
run the model).
Your help would be greatly appreciated as always
many thanks
Peter
— In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“AnyBody Support” <support@…> wrote:
>
> Hi Peter
>
> My experience is that using trial and error mostly creates errors:D I
> think it is important that the model always runs, so start letting
> everything be “not driven” and then you add/remove a DOF once at a
> time. Depending on how you drive the model it is always good to be
> able to see the axis of the coordinate system in the reference frame
> you drive it in. It is always a good idea to write down in what DOF
> that you want to drive the different segments, in that way you can
> sort of plan your programming:D
> What model did you start from? the gait3d model?
>
> Hope you get it right:D else please post again.
>
> Best regards
> Christian, AnyBody Support
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“peter.worsley” <peter.worsley@>
> wrote:
> >
> > Hi everyone,
> >
> > I’ve been collecting gait data for my project and trying to run
> > kinematic analysis. i have been setting up my joints and drivers
> > folder to constrain/drive the model. I am finding it hard to visualise
> > how to drive/constrain each degree of freedom. As a result i am using
> > trial and error in order to try and run the model. Is there any
> > function that allows me to check what plane a DOF needs to be
driven in?
> >
> > many thanks
> >
> > Peter
> >
>
[Non-text portions of this message have been removed]
Hi Soren,
I took your advice and tried to run the C3D data through the
freeposture model. I have been putting some marker drivers into the
pelvis and i have now reached the point where there are no kinematic
indeterminancies, and I feel the markers are driving in the right
directions. But i’m still getting the following multiple complaints,
here’s an example;
Constraint no. 127 above error tolerance 0.000001, error = 0.000001. -
constr. #1 in
‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
Constraint no. 128 above error tolerance 0.000001, error = 0.000001. -
constr. #2 in
‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
ERROR(OBJ.MCH.KIN3) : C:\Documents and
Settings\prw204\Desktop\PhD\Anybody\Repository.6.1\ARep\Aalborg\FreePosture107\F
reePosture.Main.any(89)
: ‘Study’ : Kinematic analysis failed in time step 0
I have been through the marker set matching the markers to the nodes
on the segments. They are as close as i can make them for the initial
position. Do i have to achive an error less than 0.0000001?
Or does this error mean something different?
Your help is always greatly appreciated
many thanks
Peter
— In anyscript@yahoogroups.com, “AnyBody Support” <support@…> wrote:
>
> Hi Peter
>
> If you think that the problem is related to the pelvis I suggest
that you
> insert joint drivers for the hip, knee and ankle joint and only
keep the
> pelvis drivers, if it still does not run and the pelvis has drivers
for six
> dof then the problem is related to driving the pelvis, and you have
isolated
> the problem.
>
> If this does not work then in line with what Christian wrote i would
suggest
> the following step procedure:
>
> 1: Take the freeposture model replace the bodymodel with the twolegs
model
> remove the unwanted drivers. This should give you a model where all
joints
> are driven.
>
> 2: Scale the model so the size is fitting place markers on the right
> location of the bones.
>
> 3: By making changes to the mannequin.any file, position the model
so that
> it fit in the best possible way with the marker locations of the
recorded
> markers. This will give you a good initial position guess which is
important
> for marker driven model like this one. 3: Now remove some of the drivers
> from the pelvis and replace them gradually by marker drivers, for
each step
> you do make sure the kinematics are running all the way through.
>
> The two leg gait model has the following dof’s: 6 rigidbody dof of
pelvis, 3
> hip, 1 knee, 2 ankle in total for both legs this gives 18, so you need
> exactly 18 drivers, and they must be able to drive these dofs. When you
> select which dof to driver for a specific marker it is usually a
good idea
> to select a direction which is normal to the long axis of the bone the
> marker is attached too. Here it is important to understand the
concept of
> the AnyKinLinear measure and the difference of using a Ref=0 or Ref=-1,
> please refer to the manual if needed. You will get the best solution
is you
> use Ref=0, because this makes the model independent of the
orientation of
> the subject. Sometimes I do a small stick figure sketch which shows the
> drivers as arrows pointing in the right directions according to the
segment
> coordinate system.
>
> 4: continue to remove the joint drivers replacing them by the marker
> drivers, one by one until they are all replaced.
>
> I hope this helps you move on.
>
> Best regards
>
> Søren, AnyBody Support
>
>
>
> _____
>
> From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com]
On Behalf
> Of peter.worsley
> Sent: 12 February 2008 13:21
> To: anyscript@yahoogroups.com
> Subject: [AnyScript] Re: kinematic indeterminate systems
>
>
>
> Hi Christian,
>
> I tried to take your advice and I put in AnyStdJoints to stop my DOF
> driver error for kinematic analysis. But I am still getting a
> indeterminant system. I have based my model on the Aalborg gait 3D,
> but I notice that you have added a sacral marker set in the model (i
> am using PSIS). I think the problem is lying in the pelvis, I have
> contrained 160 out of 161 contraints and tried to make a sacral marker
> but with no luck (its coming up with unsovlable constraint, when i
> run the model).
>
> Your help would be greatly appreciated as always
>
> many thanks
>
> Peter
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
> “AnyBody Support” <support@> wrote:
> >
> > Hi Peter
> >
> > My experience is that using trial and error mostly creates errors:D I
> > think it is important that the model always runs, so start letting
> > everything be “not driven” and then you add/remove a DOF once at a
> > time. Depending on how you drive the model it is always good to be
> > able to see the axis of the coordinate system in the reference frame
> > you drive it in. It is always a good idea to write down in what DOF
> > that you want to drive the different segments, in that way you can
> > sort of plan your programming:D
> > What model did you start from? the gait3d model?
> >
> > Hope you get it right:D else please post again.
> >
> > Best regards
> > Christian, AnyBody Support
> >
> > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com,
> “peter.worsley” <peter.worsley@>
> > wrote:
> > >
> > > Hi everyone,
> > >
> > > I’ve been collecting gait data for my project and trying to run
> > > kinematic analysis. i have been setting up my joints and drivers
> > > folder to constrain/drive the model. I am finding it hard to
visualise
> > > how to drive/constrain each degree of freedom. As a result i am
using
> > > trial and error in order to try and run the model. Is there any
> > > function that allows me to check what plane a DOF needs to be
> driven in?
> > >
> > > many thanks
> > >
> > > Peter
> > >
> >
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
Hi Peter
I am a little unsure if your kin analysis did run some steps or it gave this
error in the first step?
Anyway it is not a good sign if the kinematics can not solve the constraints
down to the default level, it is usually a sign of that something is wrong
in the model. You can decrease this accuracy if you like using the variable
KinematicTol in the study, but doing so may also lead to jumpy muscle
activities because you will introduce some slack in all the constraints.
In the debug phase it can be a good idea some times to increase the
tolerance since it might make model move and you can sometimes get an idea
on what is the problem.
As I have understood the way you are driving the model you have now joint
drivers in all joints and you are using the markers for driving the pelvis
only. If this is the case I think it is a little strange that the
constraints it can not fulfil is the hipdriver, if the leg is driven by
joint drivers only, it should always be possible for it to control this
joint.
Another thing you could try is to apply the drivers one by one on the
pelvis. So first you could add a marker driver that would drive it forward
and remove the equivalent constraint in the driver between pelvis and global
ref. Then drive the height and so on. In this way it should be possible to
always have a running model.
I hope this helps you move on.
Best regards
Søren
From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 13 February 2008 18:39
To: anyscript@yahoogroups.com
Subject: [AnyScript] Re: kinematic indeterminate systems
Hi Soren,
I took your advice and tried to run the C3D data through the
freeposture model. I have been putting some marker drivers into the
pelvis and i have now reached the point where there are no kinematic
indeterminancies, and I feel the markers are driving in the right
directions. But i’m still getting the following multiple complaints,
here’s an example;
Constraint no. 127 above error tolerance 0.000001, error = 0.000001. -
constr. #1 in
‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
Constraint no. 128 above error tolerance 0.000001, error = 0.000001. -
constr. #2 in
‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
ERROR(OBJ.MCH.KIN3) : C:\Documents and
Settings\prw204\Desktop\PhD\Anybody\Repository.6.1\ARep\Aalborg\FreePosture1
07\FreePosture.Main.any(89)
: ‘Study’ : Kinematic analysis failed in time step 0
I have been through the marker set matching the markers to the nodes
on the segments. They are as close as i can make them for the initial
position. Do i have to achive an error less than 0.0000001?
Or does this error mean something different?
Your help is always greatly appreciated
many thanks
Peter
— In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“AnyBody Support” <support@…> wrote:
>
> Hi Peter
>
> If you think that the problem is related to the pelvis I suggest
that you
> insert joint drivers for the hip, knee and ankle joint and only
keep the
> pelvis drivers, if it still does not run and the pelvis has drivers
for six
> dof then the problem is related to driving the pelvis, and you have
isolated
> the problem.
>
> If this does not work then in line with what Christian wrote i would
suggest
> the following step procedure:
>
> 1: Take the freeposture model replace the bodymodel with the twolegs
model
> remove the unwanted drivers. This should give you a model where all
joints
> are driven.
>
> 2: Scale the model so the size is fitting place markers on the right
> location of the bones.
>
> 3: By making changes to the mannequin.any file, position the model
so that
> it fit in the best possible way with the marker locations of the
recorded
> markers. This will give you a good initial position guess which is
important
> for marker driven model like this one. 3: Now remove some of the drivers
> from the pelvis and replace them gradually by marker drivers, for
each step
> you do make sure the kinematics are running all the way through.
>
> The two leg gait model has the following dof’s: 6 rigidbody dof of
pelvis, 3
> hip, 1 knee, 2 ankle in total for both legs this gives 18, so you need
> exactly 18 drivers, and they must be able to drive these dofs. When you
> select which dof to driver for a specific marker it is usually a
good idea
> to select a direction which is normal to the long axis of the bone the
> marker is attached too. Here it is important to understand the
concept of
> the AnyKinLinear measure and the difference of using a Ref=0 or Ref=-1,
> please refer to the manual if needed. You will get the best solution
is you
> use Ref=0, because this makes the model independent of the
orientation of
> the subject. Sometimes I do a small stick figure sketch which shows the
> drivers as arrows pointing in the right directions according to the
segment
> coordinate system.
>
> 4: continue to remove the joint drivers replacing them by the marker
> drivers, one by one until they are all replaced.
>
> I hope this helps you move on.
>
> Best regards
>
> Søren, AnyBody Support
>
>
>
> _____
>
> From: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
[mailto:anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com]
On Behalf
> Of peter.worsley
> Sent: 12 February 2008 13:21
> To: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
> Subject: [AnyScript] Re: kinematic indeterminate systems
>
>
>
> Hi Christian,
>
> I tried to take your advice and I put in AnyStdJoints to stop my DOF
> driver error for kinematic analysis. But I am still getting a
> indeterminant system. I have based my model on the Aalborg gait 3D,
> but I notice that you have added a sacral marker set in the model (i
> am using PSIS). I think the problem is lying in the pelvis, I have
> contrained 160 out of 161 contraints and tried to make a sacral marker
> but with no luck (its coming up with unsovlable constraint, when i
> run the model).
>
> Your help would be greatly appreciated as always
>
> many thanks
>
> Peter
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
> “AnyBody Support” <support@> wrote:
> >
> > Hi Peter
> >
> > My experience is that using trial and error mostly creates errors:D I
> > think it is important that the model always runs, so start letting
> > everything be “not driven” and then you add/remove a DOF once at a
> > time. Depending on how you drive the model it is always good to be
> > able to see the axis of the coordinate system in the reference frame
> > you drive it in. It is always a good idea to write down in what DOF
> > that you want to drive the different segments, in that way you can
> > sort of plan your programming:D
> > What model did you start from? the gait3d model?
> >
> > Hope you get it right:D else please post again.
> >
> > Best regards
> > Christian, AnyBody Support
> >
> > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com,
> “peter.worsley” <peter.worsley@>
> > wrote:
> > >
> > > Hi everyone,
> > >
> > > I’ve been collecting gait data for my project and trying to run
> > > kinematic analysis. i have been setting up my joints and drivers
> > > folder to constrain/drive the model. I am finding it hard to
visualise
> > > how to drive/constrain each degree of freedom. As a result i am
using
> > > trial and error in order to try and run the model. Is there any
> > > function that allows me to check what plane a DOF needs to be
> driven in?
> > >
> > > many thanks
> > >
> > > Peter
> > >
> >
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
[Non-text portions of this message have been removed]
Hi Soren,
I have tried to change the markers one at a time, but with no success.
The model still fails to run set initial positions despite attempts
++. I think there is something fundamentally wrong with the model, and
i would like to solve the problem personally but i have not made any
progress for a while. i will post the file up on the forum, see if
anyone else has had similar problems. I will attach a word document
explaining what i have changed to the model and the recurrent
problems. The file is freeposture107.
many thanks
peter
— In anyscript@yahoogroups.com, “AnyBody Support” <support@…> wrote:
>
> Hi Peter
>
> I am a little unsure if your kin analysis did run some steps or it
gave this
> error in the first step?
>
> Anyway it is not a good sign if the kinematics can not solve the
constraints
> down to the default level, it is usually a sign of that something is
wrong
> in the model. You can decrease this accuracy if you like using the
variable
> KinematicTol in the study, but doing so may also lead to jumpy muscle
> activities because you will introduce some slack in all the
constraints.
>
> In the debug phase it can be a good idea some times to increase the
> tolerance since it might make model move and you can sometimes get
an idea
> on what is the problem.
>
> As I have understood the way you are driving the model you have now
joint
> drivers in all joints and you are using the markers for driving the
pelvis
> only. If this is the case I think it is a little strange that the
> constraints it can not fulfil is the hipdriver, if the leg is driven by
> joint drivers only, it should always be possible for it to control this
> joint.
>
> Another thing you could try is to apply the drivers one by one on the
> pelvis. So first you could add a marker driver that would drive it
forward
> and remove the equivalent constraint in the driver between pelvis
and global
> ref. Then drive the height and so on. In this way it should be
possible to
> always have a running model.
>
> I hope this helps you move on.
>
> Best regards
>
> Søren
>
>
>
>
>
>
>
>
>
>
>
>
>
> _____
>
> From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com]
On Behalf
> Of peter.worsley
> Sent: 13 February 2008 18:39
> To: anyscript@yahoogroups.com
> Subject: [AnyScript] Re: kinematic indeterminate systems
>
>
>
> Hi Soren,
>
> I took your advice and tried to run the C3D data through the
> freeposture model. I have been putting some marker drivers into the
> pelvis and i have now reached the point where there are no kinematic
> indeterminancies, and I feel the markers are driving in the right
> directions. But i’m still getting the following multiple complaints,
> here’s an example;
>
> Constraint no. 127 above error tolerance 0.000001, error = 0.000001. -
> constr. #1 in
> ‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
> Constraint no. 128 above error tolerance 0.000001, error = 0.000001. -
> constr. #2 in
> ‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
> ERROR(OBJ.MCH.KIN3) : C:\Documents and
>
Settings\prw204\Desktop\PhD\Anybody\Repository.6.1\ARep\Aalborg\FreePosture1
> 07\FreePosture.Main.any(89)
> : ‘Study’ : Kinematic analysis failed in time step 0
>
> I have been through the marker set matching the markers to the nodes
> on the segments. They are as close as i can make them for the initial
> position. Do i have to achive an error less than 0.0000001?
> Or does this error mean something different?
>
> Your help is always greatly appreciated
>
> many thanks
>
> Peter
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
> “AnyBody Support” <support@> wrote:
> >
> > Hi Peter
> >
> > If you think that the problem is related to the pelvis I suggest
> that you
> > insert joint drivers for the hip, knee and ankle joint and only
> keep the
> > pelvis drivers, if it still does not run and the pelvis has drivers
> for six
> > dof then the problem is related to driving the pelvis, and you have
> isolated
> > the problem.
> >
> > If this does not work then in line with what Christian wrote i would
> suggest
> > the following step procedure:
> >
> > 1: Take the freeposture model replace the bodymodel with the twolegs
> model
> > remove the unwanted drivers. This should give you a model where all
> joints
> > are driven.
> >
> > 2: Scale the model so the size is fitting place markers on the right
> > location of the bones.
> >
> > 3: By making changes to the mannequin.any file, position the model
> so that
> > it fit in the best possible way with the marker locations of the
> recorded
> > markers. This will give you a good initial position guess which is
> important
> > for marker driven model like this one. 3: Now remove some of the
drivers
> > from the pelvis and replace them gradually by marker drivers, for
> each step
> > you do make sure the kinematics are running all the way through.
> >
> > The two leg gait model has the following dof’s: 6 rigidbody dof of
> pelvis, 3
> > hip, 1 knee, 2 ankle in total for both legs this gives 18, so you need
> > exactly 18 drivers, and they must be able to drive these dofs.
When you
> > select which dof to driver for a specific marker it is usually a
> good idea
> > to select a direction which is normal to the long axis of the bone the
> > marker is attached too. Here it is important to understand the
> concept of
> > the AnyKinLinear measure and the difference of using a Ref=0 or
Ref=-1,
> > please refer to the manual if needed. You will get the best solution
> is you
> > use Ref=0, because this makes the model independent of the
> orientation of
> > the subject. Sometimes I do a small stick figure sketch which
shows the
> > drivers as arrows pointing in the right directions according to the
> segment
> > coordinate system.
> >
> > 4: continue to remove the joint drivers replacing them by the marker
> > drivers, one by one until they are all replaced.
> >
> > I hope this helps you move on.
> >
> > Best regards
> >
> > Søren, AnyBody Support
> >
> >
> >
> > _____
> >
> > From: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
> [mailto:anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com]
> On Behalf
> > Of peter.worsley
> > Sent: 12 February 2008 13:21
> > To: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
> > Subject: [AnyScript] Re: kinematic indeterminate systems
> >
> >
> >
> > Hi Christian,
> >
> > I tried to take your advice and I put in AnyStdJoints to stop my DOF
> > driver error for kinematic analysis. But I am still getting a
> > indeterminant system. I have based my model on the Aalborg gait 3D,
> > but I notice that you have added a sacral marker set in the model (i
> > am using PSIS). I think the problem is lying in the pelvis, I have
> > contrained 160 out of 161 contraints and tried to make a sacral marker
> > but with no luck (its coming up with unsovlable constraint, when i
> > run the model).
> >
> > Your help would be greatly appreciated as always
> >
> > many thanks
> >
> > Peter
> >
> > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com,
> > “AnyBody Support” <support@> wrote:
> > >
> > > Hi Peter
> > >
> > > My experience is that using trial and error mostly creates
errors:D I
> > > think it is important that the model always runs, so start letting
> > > everything be “not driven” and then you add/remove a DOF once at a
> > > time. Depending on how you drive the model it is always good to be
> > > able to see the axis of the coordinate system in the reference frame
> > > you drive it in. It is always a good idea to write down in what DOF
> > > that you want to drive the different segments, in that way you can
> > > sort of plan your programming:D
> > > What model did you start from? the gait3d model?
> > >
> > > Hope you get it right:D else please post again.
> > >
> > > Best regards
> > > Christian, AnyBody Support
> > >
> > > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
> ps.com,
> > “peter.worsley” <peter.worsley@>
> > > wrote:
> > > >
> > > > Hi everyone,
> > > >
> > > > I’ve been collecting gait data for my project and trying to run
> > > > kinematic analysis. i have been setting up my joints and drivers
> > > > folder to constrain/drive the model. I am finding it hard to
> visualise
> > > > how to drive/constrain each degree of freedom. As a result i am
> using
> > > > trial and error in order to try and run the model. Is there any
> > > > function that allows me to check what plane a DOF needs to be
> > driven in?
> > > >
> > > > many thanks
> > > >
> > > > Peter
> > > >
> > >
> >
> >
> >
> >
> >
> > [Non-text portions of this message have been removed]
> >
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
Hej Peter
I have done a few corrections in the model, and now it is driven in the x
direction by a marker.
The problem was in the jointanddrivers2.any file, the way the free marker
was selected
AnyRefFrame &ref2=Main.MarkerTopology.M5.Seg
is not correct it should be
AnyRefFrame &ref2=Main.MarkerTopology.M5.FreeMarker;
Before it was trying to drive the asis marker wrt. the pelvis segment which
will not work, so you have to point at the free floating marker.
I have updated the JointAndDrivers2.any file in your folder and I also
updated the MarkerMover.any file to include a guess on the initial marker
position.
Best regards
Søren, AnyBody Support
From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 18 February 2008 12:19
To: anyscript@yahoogroups.com
Subject: [AnyScript] Re: kinematic indeterminate systems
Hi Soren,
I have tried to change the markers one at a time, but with no success.
The model still fails to run set initial positions despite attempts
++. I think there is something fundamentally wrong with the model, and
i would like to solve the problem personally but i have not made any
progress for a while. i will post the file up on the forum, see if
anyone else has had similar problems. I will attach a word document
explaining what i have changed to the model and the recurrent
problems. The file is freeposture107.
many thanks
peter
— In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“AnyBody Support” <support@…> wrote:
>
> Hi Peter
>
> I am a little unsure if your kin analysis did run some steps or it
gave this
> error in the first step?
>
> Anyway it is not a good sign if the kinematics can not solve the
constraints
> down to the default level, it is usually a sign of that something is
wrong
> in the model. You can decrease this accuracy if you like using the
variable
> KinematicTol in the study, but doing so may also lead to jumpy muscle
> activities because you will introduce some slack in all the
constraints.
>
> In the debug phase it can be a good idea some times to increase the
> tolerance since it might make model move and you can sometimes get
an idea
> on what is the problem.
>
> As I have understood the way you are driving the model you have now
joint
> drivers in all joints and you are using the markers for driving the
pelvis
> only. If this is the case I think it is a little strange that the
> constraints it can not fulfil is the hipdriver, if the leg is driven by
> joint drivers only, it should always be possible for it to control this
> joint.
>
> Another thing you could try is to apply the drivers one by one on the
> pelvis. So first you could add a marker driver that would drive it
forward
> and remove the equivalent constraint in the driver between pelvis
and global
> ref. Then drive the height and so on. In this way it should be
possible to
> always have a running model.
>
> I hope this helps you move on.
>
> Best regards
>
> Søren
>
>
>
>
>
>
>
>
>
>
>
>
>
> _____
>
> From: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
[mailto:anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com]
On Behalf
> Of peter.worsley
> Sent: 13 February 2008 18:39
> To: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
> Subject: [AnyScript] Re: kinematic indeterminate systems
>
>
>
> Hi Soren,
>
> I took your advice and tried to run the C3D data through the
> freeposture model. I have been putting some marker drivers into the
> pelvis and i have now reached the point where there are no kinematic
> indeterminancies, and I feel the markers are driving in the right
> directions. But i’m still getting the following multiple complaints,
> here’s an example;
>
> Constraint no. 127 above error tolerance 0.000001, error = 0.000001. -
> constr. #1 in
> ‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
> Constraint no. 128 above error tolerance 0.000001, error = 0.000001. -
> constr. #2 in
> ‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
> ERROR(OBJ.MCH.KIN3) : C:\Documents and
>
Settings\prw204\Desktop\PhD\Anybody\Repository.6.1\ARep\Aalborg\FreePosture1
> 07\FreePosture.Main.any(89)
> : ‘Study’ : Kinematic analysis failed in time step 0
>
> I have been through the marker set matching the markers to the nodes
> on the segments. They are as close as i can make them for the initial
> position. Do i have to achive an error less than 0.0000001?
> Or does this error mean something different?
>
> Your help is always greatly appreciated
>
> many thanks
>
> Peter
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
> “AnyBody Support” <support@> wrote:
> >
> > Hi Peter
> >
> > If you think that the problem is related to the pelvis I suggest
> that you
> > insert joint drivers for the hip, knee and ankle joint and only
> keep the
> > pelvis drivers, if it still does not run and the pelvis has drivers
> for six
> > dof then the problem is related to driving the pelvis, and you have
> isolated
> > the problem.
> >
> > If this does not work then in line with what Christian wrote i would
> suggest
> > the following step procedure:
> >
> > 1: Take the freeposture model replace the bodymodel with the twolegs
> model
> > remove the unwanted drivers. This should give you a model where all
> joints
> > are driven.
> >
> > 2: Scale the model so the size is fitting place markers on the right
> > location of the bones.
> >
> > 3: By making changes to the mannequin.any file, position the model
> so that
> > it fit in the best possible way with the marker locations of the
> recorded
> > markers. This will give you a good initial position guess which is
> important
> > for marker driven model like this one. 3: Now remove some of the
drivers
> > from the pelvis and replace them gradually by marker drivers, for
> each step
> > you do make sure the kinematics are running all the way through.
> >
> > The two leg gait model has the following dof’s: 6 rigidbody dof of
> pelvis, 3
> > hip, 1 knee, 2 ankle in total for both legs this gives 18, so you need
> > exactly 18 drivers, and they must be able to drive these dofs.
When you
> > select which dof to driver for a specific marker it is usually a
> good idea
> > to select a direction which is normal to the long axis of the bone the
> > marker is attached too. Here it is important to understand the
> concept of
> > the AnyKinLinear measure and the difference of using a Ref=0 or
Ref=-1,
> > please refer to the manual if needed. You will get the best solution
> is you
> > use Ref=0, because this makes the model independent of the
> orientation of
> > the subject. Sometimes I do a small stick figure sketch which
shows the
> > drivers as arrows pointing in the right directions according to the
> segment
> > coordinate system.
> >
> > 4: continue to remove the joint drivers replacing them by the marker
> > drivers, one by one until they are all replaced.
> >
> > I hope this helps you move on.
> >
> > Best regards
> >
> > Søren, AnyBody Support
> >
> >
> >
> > _____
> >
> > From: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
> [mailto:anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com]
> On Behalf
> > Of peter.worsley
> > Sent: 12 February 2008 13:21
> > To: anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com
> > Subject: [AnyScript] Re: kinematic indeterminate systems
> >
> >
> >
> > Hi Christian,
> >
> > I tried to take your advice and I put in AnyStdJoints to stop my DOF
> > driver error for kinematic analysis. But I am still getting a
> > indeterminant system. I have based my model on the Aalborg gait 3D,
> > but I notice that you have added a sacral marker set in the model (i
> > am using PSIS). I think the problem is lying in the pelvis, I have
> > contrained 160 out of 161 contraints and tried to make a sacral marker
> > but with no luck (its coming up with unsovlable constraint, when i
> > run the model).
> >
> > Your help would be greatly appreciated as always
> >
> > many thanks
> >
> > Peter
> >
> > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com,
> > “AnyBody Support” <support@> wrote:
> > >
> > > Hi Peter
> > >
> > > My experience is that using trial and error mostly creates
errors:D I
> > > think it is important that the model always runs, so start letting
> > > everything be “not driven” and then you add/remove a DOF once at a
> > > time. Depending on how you drive the model it is always good to be
> > > able to see the axis of the coordinate system in the reference frame
> > > you drive it in. It is always a good idea to write down in what DOF
> > > that you want to drive the different segments, in that way you can
> > > sort of plan your programming:D
> > > What model did you start from? the gait3d model?
> > >
> > > Hope you get it right:D else please post again.
> > >
> > > Best regards
> > > Christian, AnyBody Support
> > >
> > > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
> ps.com,
> > “peter.worsley” <peter.worsley@>
> > > wrote:
> > > >
> > > > Hi everyone,
> > > >
> > > > I’ve been collecting gait data for my project and trying to run
> > > > kinematic analysis. i have been setting up my joints and drivers
> > > > folder to constrain/drive the model. I am finding it hard to
> visualise
> > > > how to drive/constrain each degree of freedom. As a result i am
> using
> > > > trial and error in order to try and run the model. Is there any
> > > > function that allows me to check what plane a DOF needs to be
> > driven in?
> > > >
> > > > many thanks
> > > >
> > > > Peter
> > > >
> > >
> >
> >
> >
> >
> >
> > [Non-text portions of this message have been removed]
> >
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
[Non-text portions of this message have been removed]
Hi Soren,
thank you very much for the advice, I now have a completely determinate system
with the marker drivers. Script notation may well be the death of me!! I look
forward to developing the model.
many thanks
Peter
AnyBody Support <support@anybodytech.com> wrote: Hej Peter
I have done a few corrections in the model, and now it is driven in the x
direction by a marker.
The problem was in the jointanddrivers2.any file, the way the free marker
was selected
AnyRefFrame &ref2=Main.MarkerTopology.M5.Seg
is not correct it should be
AnyRefFrame &ref2=Main.MarkerTopology.M5.FreeMarker;
Before it was trying to drive the asis marker wrt. the pelvis segment which
will not work, so you have to point at the free floating marker.
I have updated the JointAndDrivers2.any file in your folder and I also
updated the MarkerMover.any file to include a guess on the initial marker
position.
Best regards
Søren, AnyBody Support
From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of peter.worsley
Sent: 18 February 2008 12:19
To: anyscript@yahoogroups.com
Subject: [AnyScript] Re: kinematic indeterminate systems
Hi Soren,
I have tried to change the markers one at a time, but with no success.
The model still fails to run set initial positions despite attempts
++. I think there is something fundamentally wrong with the model, and
i would like to solve the problem personally but i have not made any
progress for a while. i will post the file up on the forum, see if
anyone else has had similar problems. I will attach a word document
explaining what i have changed to the model and the recurrent
problems. The file is freeposture107.
many thanks
peter
— In anyscript@yahoogrou ps.com,
“AnyBody Support” wrote:
>
> Hi Peter
>
> I am a little unsure if your kin analysis did run some steps or it
gave this
> error in the first step?
>
> Anyway it is not a good sign if the kinematics can not solve the
constraints
> down to the default level, it is usually a sign of that something is
wrong
> in the model. You can decrease this accuracy if you like using the
variable
> KinematicTol in the study, but doing so may also lead to jumpy muscle
> activities because you will introduce some slack in all the
constraints.
>
> In the debug phase it can be a good idea some times to increase the
> tolerance since it might make model move and you can sometimes get
an idea
> on what is the problem.
>
> As I have understood the way you are driving the model you have now
joint
> drivers in all joints and you are using the markers for driving the
pelvis
> only. If this is the case I think it is a little strange that the
> constraints it can not fulfil is the hipdriver, if the leg is driven by
> joint drivers only, it should always be possible for it to control this
> joint.
>
> Another thing you could try is to apply the drivers one by one on the
> pelvis. So first you could add a marker driver that would drive it
forward
> and remove the equivalent constraint in the driver between pelvis
and global
> ref. Then drive the height and so on. In this way it should be
possible to
> always have a running model.
>
> I hope this helps you move on.
>
> Best regards
>
> Søren
>
>
>
>
>
>
>
>
>
>
>
>
>
> _____
>
> From: anyscript@yahoogrou ps.com
[mailto:anyscript@yahoogrou ps.com]
On Behalf
> Of peter.worsley
> Sent: 13 February 2008 18:39
> To: anyscript@yahoogrou ps.com
> Subject: [AnyScript] Re: kinematic indeterminate systems
>
>
>
> Hi Soren,
>
> I took your advice and tried to run the C3D data through the
> freeposture model. I have been putting some marker drivers into the
> pelvis and i have now reached the point where there are no kinematic
> indeterminancies, and I feel the markers are driving in the right
> directions. But i’m still getting the following multiple complaints,
> here’s an example;
>
> Constraint no. 127 above error tolerance 0.000001, error = 0.000001. -
> constr. #1 in
> ‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
> Constraint no. 128 above error tolerance 0.000001, error = 0.000001. -
> constr. #2 in
> ‘Main.Model.ModelEnvironmentConnection.Drivers.HipDriverRight’
> ERROR(OBJ.MCH.KIN3) : C:\Documents and
>
Settings\prw204\Desktop\PhD\Anybody\Repository.6.1\ARep\Aalborg\FreePosture1
> 07\FreePosture.Main.any(89)
> : ‘Study’ : Kinematic analysis failed in time step 0
>
> I have been through the marker set matching the markers to the nodes
> on the segments. They are as close as i can make them for the initial
> position. Do i have to achive an error less than 0.0000001?
> Or does this error mean something different?
>
> Your help is always greatly appreciated
>
> many thanks
>
> Peter
>
> — In anyscript@yahoogrou ps.com,
> “AnyBody Support” wrote:
> >
> > Hi Peter
> >
> > If you think that the problem is related to the pelvis I suggest
> that you
> > insert joint drivers for the hip, knee and ankle joint and only
> keep the
> > pelvis drivers, if it still does not run and the pelvis has drivers
> for six
> > dof then the problem is related to driving the pelvis, and you have
> isolated
> > the problem.
> >
> > If this does not work then in line with what Christian wrote i would
> suggest
> > the following step procedure:
> >
> > 1: Take the freeposture model replace the bodymodel with the twolegs
> model
> > remove the unwanted drivers. This should give you a model where all
> joints
> > are driven.
> >
> > 2: Scale the model so the size is fitting place markers on the right
> > location of the bones.
> >
> > 3: By making changes to the mannequin.any file, position the model
> so that
> > it fit in the best possible way with the marker locations of the
> recorded
> > markers. This will give you a good initial position guess which is
> important
> > for marker driven model like this one. 3: Now remove some of the
drivers
> > from the pelvis and replace them gradually by marker drivers, for
> each step
> > you do make sure the kinematics are running all the way through.
> >
> > The two leg gait model has the following dof’s: 6 rigidbody dof of
> pelvis, 3
> > hip, 1 knee, 2 ankle in total for both legs this gives 18, so you need
> > exactly 18 drivers, and they must be able to drive these dofs.
When you
> > select which dof to driver for a specific marker it is usually a
> good idea
> > to select a direction which is normal to the long axis of the bone the
> > marker is attached too. Here it is important to understand the
> concept of
> > the AnyKinLinear measure and the difference of using a Ref=0 or
Ref=-1,
> > please refer to the manual if needed. You will get the best solution
> is you
> > use Ref=0, because this makes the model independent of the
> orientation of
> > the subject. Sometimes I do a small stick figure sketch which
shows the
> > drivers as arrows pointing in the right directions according to the
> segment
> > coordinate system.
> >
> > 4: continue to remove the joint drivers replacing them by the marker
> > drivers, one by one until they are all replaced.
> >
> > I hope this helps you move on.
> >
> > Best regards
> >
> > Søren, AnyBody Support
> >
> >
> >
> > _____
> >
> > From: anyscript@yahoogrou ps.com
> [mailto:anyscript@yahoogrou ps.com]
> On Behalf
> > Of peter.worsley
> > Sent: 12 February 2008 13:21
> > To: anyscript@yahoogrou ps.com
> > Subject: [AnyScript] Re: kinematic indeterminate systems
> >
> >
> >
> > Hi Christian,
> >
> > I tried to take your advice and I put in AnyStdJoints to stop my DOF
> > driver error for kinematic analysis. But I am still getting a
> > indeterminant system. I have based my model on the Aalborg gait 3D,
> > but I notice that you have added a sacral marker set in the model (i
> > am using PSIS). I think the problem is lying in the pelvis, I have
> > contrained 160 out of 161 contraints and tried to make a sacral marker
> > but with no luck (its coming up with unsovlable constraint, when i
> > run the model).
> >
> > Your help would be greatly appreciated as always
> >
> > many thanks
> >
> > Peter
> >
> > — In anyscript@yahoogrou
ps.com,
> > “AnyBody Support” wrote:
> > >
> > > Hi Peter
> > >
> > > My experience is that using trial and error mostly creates
errors:D I
> > > think it is important that the model always runs, so start letting
> > > everything be “not driven” and then you add/remove a DOF once at a
> > > time. Depending on how you drive the model it is always good to be
> > > able to see the axis of the coordinate system in the reference frame
> > > you drive it in. It is always a good idea to write down in what DOF
> > > that you want to drive the different segments, in that way you can
> > > sort of plan your programming:D
> > > What model did you start from? the gait3d model?
> > >
> > > Hope you get it right:D else please post again.
> > >
> > > Best regards
> > > Christian, AnyBody Support
> > >
> > > — In anyscript@yahoogrou
> ps.com,
> > “peter.worsley”
> > > wrote:
> > > >
> > > > Hi everyone,
> > > >
> > > > I’ve been collecting gait data for my project and trying to run
> > > > kinematic analysis. i have been setting up my joints and drivers
> > > > folder to constrain/drive the model. I am finding it hard to
> visualise
> > > > how to drive/constrain each degree of freedom. As a result i am
> using
> > > > trial and error in order to try and run the model. Is there any
> > > > function that allows me to check what plane a DOF needs to be
> > driven in?
> > > >
> > > > many thanks
> > > >
> > > > Peter
> > > >
> > >
> >
> >
> >
> >
> >
> > [Non-text portions of this message have been removed]
> >
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
[Non-text portions of this message have been removed]
Users of the AnyBody Modeling System help each other create biomechanical models
in the AnyScript language.
Yahoo! Groups Links
Yahoo! Answers - Get better answers from someone who knows. Tryit now.
[Non-text portions of this message have been removed]