[ODE] friction and restitution

[Erin Catto]

No such formulas exist because friction and restitution are only defined for
pairs of bodies, not for bodies in isolation.

Be careful not to use ODE for engineering purposes because the contact model
is of low accuracy, suitable only for entertainment purposes. If you need an
accurate result, use an engineering grade finite element package.

If accuracy is not important to you, then just use a simple formula, like
the average.

Erin

-----Original Message-----
From: ode-bounces at q12.org [mailto:ode-bounces at q12.org] On Behalf Of Dimitris
Papavasiliou
Sent: Saturday, October 02, 2004 3:29 PM
To: ode at q12.org
Subject: Re: [ODE] friction and restitution

On Sunday 03 October 2004 00:45, Adam D. Moss wrote:
> Dimitris Papavasiliou wrote:
> > So isn't there a way to define some sort of friction and restitution
> > coefficients *per body* not per pair and then be able to calculate the
> > coefficients for the pair from them?
>
> Certainly.  Attach your own structure with this (and any other)
> per-body attributes to the body with dBodySetData().  Then in
> the collision callback, use dBodyGetData() to get the attributes
> off the bodies again and calculate appropriate friction etc.
> for the pair.

Err, no, this was supposed to be a physics related question, not an ODE 
question. The question was whether such attributes exist in the first place,

that is whether some measure of friction and restitution can be defined for
a 
single body (from which we can calculate the respective (approximate)
measure 
for a given pair) and which attributes these are.

For example: if such a friction attribute exists and for wood it is u1 and
for 
steel u2 then the friction coefficients for wood-wood, wood-steel, 
steel-steel should be f(u1, u1), f(u2, u2), f(u3, u3) respectively, where f 
is some function. I've been searching all over the web for something like 
that but to no avail.

Dimitris
_______________________________________________
ODE mailing list
ODE at q12.org
http://q12.org/mailman/listinfo/ode