[ODE] Torque due to friction.

[Thomas Harte]

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

--=_NextPart_Caramail_0321841032545276_ID
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit

> If my object is a perfect sphere, how is the amount 
> of angular velocity (torque) and linear velocity due to friction 
> calculated?
> 
> What are the parameters 
> involved?

I don't know about the specifics of ODE, but I can explain the maths in an abstract form.

Friction in ODE follows the Coulomb model, which means F <= mu*R. R is the resistive 
force that is preventing your object from falling through the surface. mu is some 
coefficient of friction that you provide. The inequality is because friction may never 
actually add energy.

A force acting at a point P on a rigid body has, as you say, a linear and an angular effect.

The linear effect is described by Newton, i.e. F=ma. So the acting force is divided by 
object mass and fed into the various integrators to affect position.

The angular effect is also fairly simple. The vector generated by PxF (i.e. the 
cross/vector product of the vector to the point of contact and the force at the contact) 
forms the angular acceleration caused by that particular contact.

For this purpose, angular position is defined to be a 3 component vector, where 
orientation is defined to be a rotation around the axis defined by the direction of the 
vector by 'the length of the vector' radians.

ODE supports a slightly more complicated system where it is possible to define the 
coefficient of friction as acting differently in two perpendicular directions. In which case 
the velocity of the body in either direction is calculated (using the dot product usually), 
then the frictional force for either direction is calculated, and summed to get total friction 
due to that constant.

Hope I made sense!

-Thomas
______________________________________________________
Check out all the latest outrageous email attachments on the Outrageous Email Chart! - http://viral.lycos.co.uk	


--=_NextPart_Caramail_0321841032545276_ID--