Re: [Algorithms] Using RAPID within a rigid body simulator

[Pierre T. <p.t...@wa...>]

> A potential problem with RAPID is that it works from a 'polygon soup', and
> as such has no concept of solidity to distinguish between insideness and
> outsideness.

Sure, but the concept does not exist in RAPID for a good reason : it does
not exist at all for polygon soups. It remains the only library so far to
deal with such cases. Tell me if I'm wrong, but say I want to simulate 256
stupid teapots, what choices do I have ?


> So, depending on how small or fast your objects are, RAPID might not be
> able to detect all your collisions (i.e. object A fully inside object B)

That's what swept volumes are for, I guess. Painful problem indeed.


> unless it's enhanced to do so. I guess this is probably not a problem
> in general, as you're probably running your physical simulation at a
fairly
> high frame rate, but it could be an issue in some cases. Of course, the
> triangle-triangle test could of course be extended to handle velocity,
> which would fix this problem.

Yes, this is the big issue: 4D collision detection. A real PITA:
- difficult to implement and visualize
- difficult to design since it suddenly links a perfectly static stand-alone
collision detection library to the heart of your dynamic simulation.

Perfectly painful.


> Also, I'm no expert on physical simulation, but one important aspect is
> computing the contact manifold (the set of points of contact) and again
> this is something RAPID would have to be enhanced with, to be able to
> provide.

Actually that's what we were discussing, I think. (...err I hope... because
that was my question at first !) The way I see it, the closest pair of
points becomes the points of contacts, as soon as the distance between the
two bodies is less than your favorite threshold.

All that junk starts to be very clear IMHO. The only nasty remaining part
will probably be the linear programming one for resting contacts, but I'm
really not in a hurry for that one.

...hmm and there are also those KDS thingies I still must read about... oh,
well.

Pierre