Forums Anyscript.org  

Go Back   Forums Anyscript.org > Professional Forum > Debug model

Notices

Reply
 
Thread Tools Display Modes
  #1  
Old 07-11-2010, 05:48 AM
cathycug cathycug is offline
Member
 
Join Date: Feb 2010
Posts: 64
Default Help with StandingLift model

Hey, AnyBody,

I am using AMMR 1.2.

Based on the example of StandingLift model, I want to make a little changes to the human body posture when lifting, with the box exactly in front of the body, like the attached figure (1).

However, after I changed the mannequin.any, and CoM Drivers, and run the initialposition, I can only get the figure (2), the box was still put at the right side of the body. I tried to input many different values in the CoM driver, but failed to get what I want:

AnyKinEqSimpleDriver
CoMDriver = {
AnyKinCoM CoM = {
AnyFolder &Body = Main.Model.HumanModel;
AnyFolder &Environment = Main.Model.EnvironmentModel;
};
MeasureOrganizer = {0,2}; // Only the x and z directions
DriverPos = {0.1, -0.05}; //Possibly I changed the values inside a little bit
DriverVel = {0.0, 0.1};
Reaction.Type = {Off,Off};
};

I also tried to use FreePostureMove (without using the CoM driver) to do the same lifting task, also failed, and got the figure (3), with two arms inside the thorax, when running initialposition. Please find all the figures attached.

Could you please advise me how to get what I want when lifting the box, when running the initial position?

Thanks for your help!

Cathy
Attached Files
File Type: zip Figures of Lifting.zip (274.3 KB, 17 views)
Reply With Quote
  #2  
Old 07-13-2010, 10:07 AM
sylvaincarbes's Avatar
sylvaincarbes sylvaincarbes is offline
AnyBody Support
 
Join Date: Mar 2009
Posts: 481
Default

Hi Cathy,

The box has its own driver for translation and rotation in the global reference frame (you can find it in the JointsAndDrivers file). Notice that in this case the CoM driver does not affect the box because its motion is already fully defined.
So if you want the box to be always exactly in front of the body you can drive the box to a fixed value in the thorax reference frame for example.
That should do what you want.

Best regards, Sylvain.
__________________
Sylvain Carbes, AnyBody Support

Please provide AMMR and AMS version numbers in case you send a debug model. The folder structure should follow the standard in the AMMR. This makes it a lot easier for us to answer your questions faster.

Consider to subscribe to a forum, click the 'ForumTools' link above the list of threads then select 'Subscribe to Forum'.
Reply With Quote
  #3  
Old 07-14-2010, 05:04 AM
cathycug cathycug is offline
Member
 
Join Date: Feb 2010
Posts: 64
Default

Hi, Sylvain and AnyBody,
Thanks for your timely response.
Actually I was able to lift the box from waist to shoulder level, with the control from box driver data. I also changed a little bit on the CoM DriverPos and Mannequin.any files.
It worked fine.

Now I want to lift the box from the floor to shoulder level, but failed in the time step 0, even I changed the data file of box driver, r0 of box, mannequin.any, and CoM driverPos.

I attached my lifting model.Could you please give me some advice how to make the model work forward?

Much appreciated for your help!

Cathy
Attached Files
File Type: zip StandingLift_floor to shoulder.zip (156.0 KB, 14 views)
Reply With Quote
  #4  
Old 07-15-2010, 11:56 AM
sylvaincarbes's Avatar
sylvaincarbes sylvaincarbes is offline
AnyBody Support
 
Join Date: Mar 2009
Posts: 481
Default

Hi Cathy,

The value for the CoMDriver in the y direction is much too low. Try with this DriverPos = {0.1, 1};

Also the position of the hands on the box is not correct, you should tilt them a little. Try to give a values 0f -50 degrees in the RHBoxDriver and LHBoxDriver.

The model will run with those changes, but you will see that the motion is not realistic. I think it is mainly due to contraining the CoM in the Y direction.

Best regards, Sylvain.
__________________
Sylvain Carbes, AnyBody Support

Please provide AMMR and AMS version numbers in case you send a debug model. The folder structure should follow the standard in the AMMR. This makes it a lot easier for us to answer your questions faster.

Consider to subscribe to a forum, click the 'ForumTools' link above the list of threads then select 'Subscribe to Forum'.
Reply With Quote
  #5  
Old 07-15-2010, 11:45 PM
cathycug cathycug is offline
Member
 
Join Date: Feb 2010
Posts: 64
Default

Sylvain,

Thanks for your response. Yes, I tried the values, and the model was unrealistic.

I tried this way: constrain the CoM driver in x and z directions, like the original one. I spent the whole day on the model, but failed again.
I am attaching the modified model, in need of your advice. The final kinematic error of the model is 0.322, which was the least error that I could reach.

Is there an easy way to make the box lifted from the floor to overhead level? Why I can be able to lift the box from waist level to overhead, while there were so many problems when lifting it from lower level?

Please advise,

I appreciate your helping me!

Cathy
Attached Files
File Type: zip StandingLift_floor to shoulder1.zip (78.1 KB, 23 views)
Reply With Quote
  #6  
Old 07-20-2010, 11:11 AM
sylvaincarbes's Avatar
sylvaincarbes sylvaincarbes is offline
AnyBody Support
 
Join Date: Mar 2009
Posts: 481
Default

Hi Cathy,

Some drivers are conflicting in the model. If you drive the CoM in X and Z you need a free dof on both directions. For X there is the ankle flexion but there is none for Z, so you need to free the thorax lateral bending for example instead of the thorax flexion/extension. It is extremely important to have a good overview of the drivers and dof. You can make small sketches to help you visualize it. Without this you may end up modifying the drivers randomly with little chance to achieve good results.

You will see that even after exchanging thorax lateral bending and flexion/extension, the model still does not look completly right. Lifting a box from floor to head is more complex than from waist to head. That last only involves motion in the arm. Whereas lifting from the floor normaly involves motion in all the body with extension of the ankles, knee, hip and trunk. Try to observe someone doing the task and then reproduce the joint motion in the model.

Best regards, Sylvain.
__________________
Sylvain Carbes, AnyBody Support

Please provide AMMR and AMS version numbers in case you send a debug model. The folder structure should follow the standard in the AMMR. This makes it a lot easier for us to answer your questions faster.

Consider to subscribe to a forum, click the 'ForumTools' link above the list of threads then select 'Subscribe to Forum'.
Reply With Quote
  #7  
Old 12-11-2017, 11:14 AM
sohane123 sohane123 is offline
Junior Member
 
Join Date: Apr 2017
Posts: 11
Default Request

Sir,
Please Reply me of my question.
Reply With Quote
  #8  
Old 12-11-2017, 12:31 PM
pgalibarov's Avatar
pgalibarov pgalibarov is offline
Senior Member
 
Join Date: Oct 2010
Posts: 926
Default

Hi Sohane,

I don't see any unanswered questions by you.

Secondly, you need to provide information about yourself in order for us to help you. We need to know who we are helping to. This is a professional forum.

Kind regards,
Pavel
Reply With Quote
  #9  
Old 12-11-2017, 12:52 PM
sohane123 sohane123 is offline
Junior Member
 
Join Date: Apr 2017
Posts: 11
Default Request

Sir,

My problem is that, your (SitToStand_Exo) is running itself. even if i given knee posvel=0. what should i do for running it different knee angle
Reply With Quote
  #10  
Old 12-11-2017, 03:26 PM
pgalibarov's Avatar
pgalibarov pgalibarov is offline
Senior Member
 
Join Date: Oct 2010
Posts: 926
Default

Please update your information for further support.

The knee flexion is not driven through explicit kinematic drivers. It is governed by the CoM motion.

Regards,
Pavel
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +2. The time now is 10:21 PM.