[ODE] O.D.E 'Thread' -> High Rate Thread/Rope Simulation

[Manohar B.S]

Hi Tanguy,
           Thanks for the comments. Presently I am
working on expanding Cylinders. The videos that I took
are running with fixed cylinders only. I did play a
lot with ERP and CFM params of ODE to get the right
flavor.

 There is a multi-rate processing to match the
frequencies of physics and haptics.

 There's still a lot to put in some knot theory,
basically to detect the knots so that I can do some
neat trick to avoid interpenetration and joint
failure.

-- Regards!
Manohar B. Srikanth
http://ssl.serc.iisc.ernet.in/~manohar



--- Tanguy Fautre
<tanguy.fautre at spaceapplications.com> wrote:

> Impressive videos!
> 
> 
> I was wondering what primitives did you use to model
> the rope. Do you 
> use a serie of cubes, spheres or cylinders? Or do
> you have another approach?
> 
> PS: nice research webpage; too bad there are no
> source code. ;-)
> 
> 
> Cheers,
> 
> Tanguy
> 
> 
> Manohar B.S wrote:
> > Hi,
> >    I used O.D.E as foundation to develop a 'thread
> /
> > rope' which can knot. The simulation uses a
> > haptic device to control one end point of the
> thread
> > to produce force feedback (Using dFixedJoint
> > constraint / Spring damper).
> > 
> > The videos/images are here:
> > 
> > 
> > On a Dual Xeon machine the physics frame rate is
> about
> > 300Hz. That's a lot.
> > 
> >  Rewriting some part of the ODE routines with
> assembly
> > language (MMX) will certainly boost the frame
> rate.
> > Especially matrix operations.
> > 
> >  Best Wishes & Many Thanks to the ODE community.
> > Manohar B. Srikanth
> > Indian Institute of Science(IISc)
> > Bangalore, INDIA.
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
> 



		
____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs