[ODE] solving penetrations "statically"

Nguyen Binh ngbinh at gmail.com
Wed Jun 7 12:47:21 MST 2006

Velocity constraint ALONE cannot solve drifting problem.
You will end up with penetration even if the constraint is satisfied. It's
true that velocity constraint can be obtained by taking derivitaives of
position constraint but you loose precision.
Also, the non-penetration constraint naturally should be stated in position
not velocity.
I don't know what's Erin non-penetration constraint but if he wrote it in
velocity level then he may need "error correction" term to preserve position
from drifting too far.

On 6/7/06, Dirk Gregorius <dirk at dirkgregorius.de> wrote:
> >> You can't do that if your nonpenetration constraint equations are
> expressed
> >> in velocity like ODE. You always have some drift in position.
> You can't do what? The method of Erin is first correct - see for example
> the thesis of Bradley Cline - and
> secondly works perfectly in practice. Especially in the context of game
> development. Each position constraint
> imposes a (hidden) velocity and acceleration constraint to the system.
> If you have a velocity constraint you also
> have geometric constraint as well (I must admit that I am not sure for
> friction constraints), but definitely for
> non-penetration constraints and joints.
> -Dirk
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode

Binh Nguyen
Computer Science Department
Rensselaer Polytechnic Institute
Troy, NY, 12180
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://q12.org/pipermail/ode/attachments/20060607/0f2c4632/attachment.htm

More information about the ODE mailing list