Continuous collision detection for deformable objects using permissible clusters

Abstract

In this paper, we propose a new data structure to perform continuous collision detection (CCD) for deformable triangular meshes. The critical component of this data structure is permissible clusters. At the preprocessing phase, the triangular meshes are divided into permissible clusters. Then, the features of the triangular meshes are assigned to the permissible clusters. At the runtime phase, the potentially colliding feature pairs are collected and they are processed only once in the elementary processing. Our method has been integrated with a normal cone-based method and compared with other CCD methods. Experimental results show that our method improves the overall performance of CCD for deformable objects.