[ODE] Silly question, Is ODE based on accurate physics ?

[Hugo Veiga]

Hello 

I have been using a robot simulation software WEBOTS for building and
testing robots, it uses ODE underneath. 

I model my omni robot's Swedish wheel with a cylinder. But the problem is
that because the simulator uses the generic attrition cone for friction I
end up having problems with wheels introducing friction to each other. I
therefore must internally modify ODE parameters namely perhaps dContactMu2,
dContactFDir1, dContactMotion1/2, etc.

I wonder how did others implement Swedish wheels?

I'm new to ODE, so can anyone tell me give me some pseudocode where to start
from?

Thank you very much.

Hugo Veiga

-----Original Message-----
From: ode-bounces at q12.org [mailto:ode-bounces at q12.org] On Behalf Of matt
stephure
Sent: Friday, February 11, 2005 03:37
To: ode at q12.org
Subject: Re: [ODE] Silly question, Is ODE based on accurate physics ?

I work for a machine learning center at the University of Alberta
(http://www.aicml.ca). We've been building up a generic robot
simulation/test suite. There's a lot of work with games, and robotics
here, including building teams for the RoboCup competitions.

Our simulator can handle OmniRobots and as of last week, finally,
Segway RMPs. If you want to shorten your code-run-debug cycle for
robotic work, ODE is a great tool to help you do that. It's not
perfect, but very useful.

I'm not sure I'd want to construct a bridge with it, but for research?
It's certainly accurate enough.



On Thu, 10 Feb 2005 23:54:21 -0000, Chris Royce <ode at croyce.co.uk> wrote:
> That could be a problem I wanted to use to this model a kangraoo as in
real
> life modelling.
> 
> What techniques did you use for modelling your robot ?
> 
> ----- Original Message -----
> From: "Patrick Barnes" <mrtrick at gmail.com>
> To: <ODE at q12.org>
> Sent: Thursday, February 10, 2005 10:29 PM
> Subject: Re: [ODE] Silly question, Is ODE based on accurate physics ?
> 
> > No, not quite.
> > It's not suitable for accurate mechanical simulation work, there are a
> > some 'shortcuts' that have been taken to make it fast enough for games
> > and other applications where perfect accuracy is not important.
> >
> > Robot development... it could be suitable. It just depends on what
> > exactly you want to simulate. I've actually built a balancing robot
> > last year for Uni (like a segway), and simulating it in ODE would I
> > think have been quite possible (damn, wish I had thought of that last
> > year).
> >
> >
> > P.S. What's with this mailing list, why isn't reply-to set to
ode at q12.org
> > ?
> >
> >
> > On Thu, 10 Feb 2005 22:15:58 -0000, Chris Royce <ode at croyce.co.uk>
wrote:
> >>
> >> Hello there, it just dawned on me, is ODE based on real time physics ?
> >>
> >> I know it is a physics engine, but does it have real time accuracy ?
> >>
> >> I see it has been used for games but would this have mechanical
> >> application
> >> for robot development etc?
> >>
> >> Cheers
> >>
> >> Chris
> >>
> >> I do apologise if this appears an obvious answer.
> >>
> >> _______________________________________________
> >> ODE mailing list
> >> ODE at q12.org
> >> http://q12.org/mailman/listinfo/ode
> >>
> >>
> >>
> >
> >
> > --
> > MrTrick
> >
----------------------------------------------------------------------------
---------
> > Catapultum habeo. Nisi pecuniam omnem mihi dabis, ad
> > caput tuum saxum immane mittam
> >
----------------------------------------------------------------------------
---------
> > _______________________________________________
> > ODE mailing list
> > ODE at q12.org
> > http://q12.org/mailman/listinfo/ode
> >
> 
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
> 


-- 
| Matt Stephure
| Software Developer
| Alberta Ingenuity Centre for Machine Learning
| stephure at gmail.com
+----------------------------------------------------------
_______________________________________________
ODE mailing list
ODE at q12.org
http://q12.org/mailman/listinfo/ode