Efficient Collision and Self-Collision Detection for Humanoids Based on Sphere Trees Hierarchies

Abstract

We present an algorithm for computing compact sphere tree hierarchies over general 3D polygonal meshes for the purpose of efficient collision detection and minimum distance computation. The sphere tree construction method we use optimizes the location of the sphere centers and radii at each level of the hierarchy to compactly contain the underlying geometry. We enhance the hierarchy construction so that it can be applied to both self-collision and obstacle collision detection for articulated robots. Finally, we introduce distance range bounds that are effective for minimizing the overall number of required distance computations. The result is a faster algorithm that performs particularly well on collections of multiple mesh objects.