[ODE] trimesh-box collisions
shalinor at gmail.com
Mon Apr 25 13:27:32 MST 2005
What's your step size? And how many contacts are you generating?
(try 100hz step size, and make sure you're generating at least 24 contacts)
On 4/25/05, j <ode at dynamica.org> wrote:
> Hi all,
> I'm trying to get my mind around the trimesh.
> but I keep having some trouble getting trimesh-box
> collisions right. There's an image at
> I do believe Ive set my environment correctly
> by using dSINGLE-compiled ODE with my own
> typedefs aligned to ODE's types.
> When I replace the trimesh by a box, everything
> works just fine; even though I do believe to have
> built the trimesh properly:
> dGeomTriMeshDataBuildSimple( tri_data, (const scalar *)&(ctrl_pnts), \
> num_vertices, (Index*)&surface_indices, num_indices );
> having num_vertices dVector3's in ctrl_pnts and the right number of
> indices. and having all triangles oriented ccw by using the correct indexing
> Whenever a box collides with the tri-mesh, it seems to collide only with the
> edges of the tri-mesh, boxes can (sometimes) fall right through the
> triangles, and
> some boxes never come to rest on the trimesh at all; they keep moving up and
> through triangles, like the ones in the image.
> I finally believe to have set the CFM and ERP values correctly to -1e-5 and
> 0.2 resp. but haven't got a clue what's wrong.
> anyone ?
> ODE mailing list
> ODE at q12.org
More information about the ODE