Jeff Reid wrote:
> > The car is barely moving sideways, but the slip angle is still +90 degrees,
> > so you get a kick back the opposite direction, now it's -90 degrees
> > slip angle, so you get another opposite kick. Back and forth, back and
> > forth...
>
> Seems a better model is needed. Maybe one based on strain, contact patch
> displacement, to determine force generated by the tire. This approach would
> be independent of speed

Right, that's what the relaxation approach does. The slip angle
doesn't just go right to 90 even though literally, according to the
definition of slip angle, it is really 90 degrees. Instead, it has to
travel a little distance before the state variable version of the slip
angle hits it. In the mean time, the force builds up gradually until
you hit an equilibrium. That's the idea, anyway. It can be a bit
tough to really make it work in practice though because you still need
just the right amount of damping to get it to work right.

The displacement stuff you mentioned is indeed a way to do it as well.
A simpler way would be to use a spring model at low speeds or when
stopped. A long time back I modelled the tire as a torsion spring at
low speeds. This worked quite well actually in the longitudinal
direction. The tire would spring back just a touch after a quick stop.
That didn't do anything for the lateral direction, but it improved the
car a lot when trying to stop on hills/banks and so on.

Todd Wasson
www.PerformanceSimulations.com
www.VirtualRC.com
http://www.PerformanceSimulations.co...ToddSim17a.wmv