A Multiresolution Discrete Element Method for Triangulated Objects with Implicit Time stepping

Simulations of many rigid bodies colliding with each other sometimes yield particularly interesting results if the objects differ significantly in size and are nonspherical. The most expensive part within such a simulation code is collision detection. We propose family novel multiscale detection algorithms that can be applied to triangulated explicit implicit time stepping methods. They well suited handle cannot represented by analytical shapes or assemblies objects. Inspired multigrid methods adaptive mesh refinement, we determine points iteratively over resolution hierarchy combine functional minimization plus penalty parameters actual comparision-based geometric distance calculation. Coarse surrogate geometry representations identify “no collision” scenarios early on otherwise an educated guess which triangle subsets next finer level might collisions. prune search tree furthermore feed conservative contact force estimates into iterative solve behind stepping. Implicit nonanalytical often prohibitive high compute cost for body simulations. Our approach reduces object-object comparison algorithmically one two orders magnitude. It also exhibits vectorization efficiency due its nature.