FAQ  •  Login

using DPI for driving constraint

<<

f13-759-gsubramani

Newbie
Newbie

Posts: 27

Joined: Mon Sep 09, 2013 9:12 am

Unread post Wed Oct 19, 2016 2:53 pm

using DPI for driving constraint

For HW 6, when I specify the pendulum angle using DP1 I have issues around 0 and pi. With the dot product, we can't resolve between positive and negative angles. Should we ignore this? I don't think this should pose a problem with kinematic constraints, especially when we just want to check for perpendicularity. The issue only occurs when we try using this as a driving constraint.
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Wed Oct 19, 2016 4:55 pm

Re: using DPI for driving constraint

Totally correct and very good interpretation of the issue at hand.
Dealing with this in a comprehensive way is not trivial.
Keep following the course you're following, I hope we're not going to hit a wall going down this path.

Guru, do you miss me?
'Cause i miss you - haven't seen you this week...
Dan
<<

michaeltaylor

Newbie
Newbie

Posts: 27

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Oct 19, 2016 8:57 pm

Re: using DPI for driving constraint

Guru,

Luning and I were discussing this topic after class today. If you get some odd results with using the DP1 constraint as a kinematic driver, it could be that you’ve hit a bifurcation point where you have two equally “good” solutions from a non-linear solver perspective (remember our friend the Implicit Function Theorem and when it holds). If that is the case, you’ll need to get creative. See the ME451 slides for more background details on singular configurations (http://sbel.wisc.edu/Courses/ME451/2014/Lectures/lecture1023.pdf).

Mike
<<

f13-759-gsubramani

Newbie
Newbie

Posts: 27

Joined: Mon Sep 09, 2013 9:12 am

Unread post Thu Oct 20, 2016 1:57 pm

Re: using DPI for driving constraint

Hi Dan,
Thanks for the reply. I'll keep doing what I am doing right now.

I've been keeping up with the online recordings of the lectures so I don't feel left out. I've had a long summer so I decided to take a break. The lectures are much more enjoyable in person though, but I'll be back on Monday!
Thanks!

Hi Mike,
Right now, I'm using matlab's fsolve to compute the solution. I need to explore what happens near 0 and pi. I make sure the initial starting point for fsolve is the previous iteration's solution, so my program doesn't converge to alternate solution that don't make sense looking at it from a time series perspective. There should be a better way of forcing this, but I don't know how.

Guru

Return to ME751 Fall 2016

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.