FAQ  •  Login

Assignment 8 (posted on March 11)

<<

ME751Anne

ME751 Student

Posts: 28

Joined: Tue Mar 02, 2010 10:14 pm

Unread post Fri Mar 12, 2010 2:21 pm

Assignment 8 (posted on March 11)

now there is a forum for HW8
Last edited by HammadM on Sat Mar 13, 2010 9:35 pm, edited 1 time in total.
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Fri Mar 12, 2010 6:34 pm

Re: Assignment 8 Questions

I plan to use the last 20 minutes on Tu to answer any questions you might have about the code component of the HW. 
dan
<<

ME751Anne

ME751 Student

Posts: 28

Joined: Tue Mar 02, 2010 10:14 pm

Unread post Sat Mar 13, 2010 8:00 pm

Re: Assignment 8 Questions

Dan, you ask for the position and velocity of point Q at the end of the pendulum. This where the pin is so the position and velocity is constantly 0. Is this what you are looking for?
Last edited by ME751Anne on Sat Mar 13, 2010 9:56 pm, edited 1 time in total.
<<

ME751Anne

ME751 Student

Posts: 28

Joined: Tue Mar 02, 2010 10:14 pm

Unread post Sat Mar 13, 2010 9:40 pm

Re: Assignment 8 (posted on March 11)

I'm having a problem with a singular Jacobian matrix when I try to do a position analysis on my pendulum. Any ideas how to debug this?

I think I overconstrained my problem.
Last edited by ME751Anne on Sat Mar 13, 2010 11:34 pm, edited 1 time in total.
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Mon Mar 15, 2010 7:37 am

Re: Assignment 8 Questions

ME751Anne wrote:Dan, you ask for the position and velocity of point Q at the end of the pendulum. This where the pin is so the position and velocity is constantly 0. Is this what you are looking for?


yes, that's exactly it.
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Mon Mar 15, 2010 7:42 am

Re: Assignment 8 (posted on March 11)

ME751Anne wrote:I'm having a problem with a singular Jacobian matrix when I try to do a position analysis on my pendulum. Any ideas how to debug this?

I think I overconstrained my problem.


Anne - overconstraining the Jacobian can't lead to a singular matrix.  If you do, you'll have a non-square matrix, which can't be singular (yet it can be of reduced row-rank).
If you do things right you should good a healthy Jacobian.
Dan
<<

ME751Anne

ME751 Student

Posts: 28

Joined: Tue Mar 02, 2010 10:14 pm

Unread post Mon Mar 15, 2010 11:18 am

Re: Assignment 8 (posted on March 11)

It says it was singular to a working precision. To get around this I just used pinv (the pseudoinverse), and no problems arise now.
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Mon Mar 15, 2010 1:46 pm

Re: Assignment 8 (posted on March 11)

ME751Anne wrote:It says it was singular to a working precision. To get around this I just used pinv (the pseudoinverse), and no problems arise now.


Anne - there is a problem if you choose the axes for the motion to coincide at time t=0.  It's too long to explain (I'm working on tomorrow's notes) but the idea is to have for the DP1 motion one axis along the global OY axis, and another axis along the O'x' axis of the pendulum.  The Jacobian should be healthy and there should be no need for the pseudoinverse.
Jim came over for office hours today and explained to him.  He said he was going to post a more elaborate explanation of what's going on.
If you have the time please to check this please confirm here that it solved your problem.
dan
<<

ME751Chris

ME751 Student

Posts: 55

Joined: Tue Mar 02, 2010 10:15 pm

Unread post Mon Mar 15, 2010 9:44 pm

Re: Assignment 8 (posted on March 11)

Can you give us a few code names that have done a good job so far?
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Tue Mar 16, 2010 6:53 am

Re: Assignment 8 (posted on March 11)

ME751Chris wrote:Can you give us a few code names that have done a good job so far?


MajorTom
unicorn
<<

ME751Toby

ME751 Student

Posts: 4

Joined: Tue Mar 02, 2010 10:38 pm

Unread post Tue Mar 16, 2010 3:11 pm

Re: Assignment 8 (posted on March 11)

How small should out tolerance be when using Newton method for position analysis?
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Tue Mar 16, 2010 5:48 pm

Re: Assignment 8 (posted on March 11)

ME751Toby wrote:How small should out tolerance be when using Newton method for position analysis?


You are dealing with a small problem here.  Stop when the corrections in r and p (the 2-norm) is less than 1E-7.
Please post a blurb in case you have any difficulties reaching this convergence criteria.  As a matter of fact, post a note if the convergence criteria works for you as well. 
Dan
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Tue Mar 16, 2010 6:27 pm

Re: Assignment 8 (posted on March 11)

The best test to verify the correctness of your results is to compute the velocity of the tip of the pendulum Q that is attached to ground through the revolute joint.
This velocity is computed as
v^Q = v + A tilde(\bar omega} (\bar s^Q)

If this is zero (or tiny), you are in great shape.
The same should hold true for the acceleration of Q, it should be zero (or some tiny value).

If you get the right answer please post a note on this discussion thread to indicate that you got it right.  Your ME751 fellows working on it might have specific questions for you.
 
Dan
<<

ME451Tyler

ME751 Student

Posts: 25

Joined: Tue Mar 02, 2010 10:19 pm

Unread post Tue Mar 16, 2010 7:03 pm

Re: Assignment 8 (posted on March 11)

I have a (close enough to) zero velocity for Q.

Any one care to share what they are getting for Q accelerations?  I'll share if you share.
Last edited by ME451Tyler on Tue Mar 16, 2010 7:05 pm, edited 1 time in total.
<<

ME751Chris

ME751 Student

Posts: 55

Joined: Tue Mar 02, 2010 10:15 pm

Unread post Tue Mar 16, 2010 7:51 pm

Re: Assignment 8 (posted on March 11)

what are people putting in for their initial r and p?
Next

Return to ME751 Spring 2010: Advanced Computational Multibody Dynamics

Who is online

Users browsing this forum: No registered users and 1 guest

cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by ST Software.