[ODE] ODE with non-penetration constraint

[Nguyen Binh]

>
> The ODE solves only at the velocity level. Note
> that even when you fully satisfied the velocity constraints, you still
> can end up with violated position constraints (e.g. because of high
> angular velocities). The ODE deals with this using Baumgarte
> stabilization. Erin Catto's has a nice presentaion about this at
> www.gphysics.com.


Because ODE solve for the contact forces so the constraints will be
satisfied at velocity level.
But in a velocity-impulse contact model will solve for contact impulses so
it will maintain the constraints *exactly* at position level and you don't
need Baumgarte stabilization which adds stiffness to the system.

-- 
--------------------------------------------------
Binh Nguyen
Computer Science Department
Rensselaer Polytechnic Institute
Troy, NY, 12180
--------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://ode.org/pipermail/ode/attachments/20070712/8b3d861c/attachment.htm