[ODE] Realistic friction based on material.

Remi Ricard remi.ricard at simlog.com
Fri May 18 09:24:28 MST 2007


> Hi.
> I would like to achieve a more "realistic" moviment of bodies, for example, a metal ball will roll faster in a stone ground than on a sand ground.
> I would like that the interation between objects is based on their materials, not on a fixed value for all.
> I think that I must check the material of each object evolved on the collision, and set the value of surface.mu, on the nearCallback function, right?

Yes this is the way to go.

> Is there any table with friction constants for most common materials?

Searching the net you will find lots tables of tables giving you the 
coefficient of friction of material A againts B.

> And havint that, how I set the correct resulting value for surface.mu? Multiply? Sum?

Since the coefficient of friction is still and empirical value (i.e. 
they don't really know what is going on at the interface). I don't think 
there is and equation to find mu for material A and B.

The coefficient of friction is always a value for 2 materials.

> Also, the way of the materials behavior deppends of their material.
> For example, if a metal ball falls on a stone ground, probably ti won't jump back, but if the ball is made of, say, plastic, it will jump several times.
> How I set this behavior?

You will have to play with all the parameter of a contact joint.


More information about the ODE mailing list