[ODE] why wheel as sphere ?

[Bob Dowland]

> Just try driving around on a TriMesh with cylinders. If you get happy

Ay there's the rub...

> > why not cylinders, that is the right shape?
	
cyl<>trimesh: cylinder rims against tri edges - lots of hard edgy contacts to snag on.

But true "inner-wheel-nature" is not in the shell used for contact detection, it's mostly about mass distribution. ie. the values you put in the inertia matrix. This is where most of the obvious dynamics props like tendency to remain vertical while spinning etc.. come from.

So when people claim they are using spheres for wheels, this may just be short-hand for we are using a spherical contact sensor/probe, others use rays of course which again aren't very wheel shaped - the devil is in the (mass) distribution.

:)

Bob.


> -----Original Message-----
> From: Michael Rauh [mailto:michael.rauh at gmail.com]
> Sent: 12 October 2004 08:48
> To: Nicola Lugato
> Cc: ode at q12.org
> Subject: Re: [ODE] why wheel as sphere ?
> 
> 
> Just try driving around on a TriMesh with cylinders. If you get happy
> with the behaviour, you can of course use them :-)
> 
> Greets
> Michael
> 
> 
> On Mon, 11 Oct 2004 20:18:11 +0200, Nicola Lugato
> <nicola.lugato at gmail.com> wrote:
> > Hi! I'm new to this list :)
> > I'm trying ODE and it seems a great library!
> > I've a question: why car simulation always uses sphere as wheel? why
> > not cylinders, that is the right shape?
> > Are there other cases where it is better to use a geom that is
> > different from the real shape ?
> > 
> > Thanks!
> > _______________________________________________
> > 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
>