[ODE] Different between velocity and position non-penetration constraints

[David Walters]

> We've made a simple demo to show the different between Stewart-Trinkle  and
> Anitescu Potra (ODE using this now) time stepper.
> As you can see, even with high time step, you'll never see penetration in ST
> time stepper.
>
> Grab the demo at my colleague's website:
> http://www.cs.rpi.edu/~sberard/lcp_applet/

Hi, thanks for the demo - if you don't mind, I have a few questions:

How feasible would it be to integrate this stepper into ODE? Would it
be possible to add in such a way as to be transparent to users who
want to remain with the existing "Anitescu Potra" method? is it
compatible with the existing joint types and collision functions?

It seems to me like a beneficial thing to add to ODE, as different
situations may yield better results with this stepper, just as they do
with the dWorldStep / dWorldQuickStep / dWorldStepFast1 choice they
have at the moment. The thought of removing object interpenetration is
very alluring!

My only worry would be code bloat, but as each stepper seems to me to
be a good example of encapsulated functionality - a good linker should
be able to discard the stepper(s) that aren't used? I wonder if this
is the case at the moment - I've never actually checked)

Regards,
Dave