Computing the Intersection-Depth of Polyhedra

Given two intersecting polyhedra P , Q and a direction d, find the smallest translation of Q along d that renders the interiors of P and Q disjoint. The same problem can also be posed without specifying the direction, in which case the minimum translation over all directions is sought. These are fundamental problems that arise in robotics and computer vision. We develop techniques for implicitly building and searching convolutions and apply them to derive efficient algorithms for these problems. 1 Introduction The computation of spatial relationships among geometric objects is a fundamental problem in such areas as robotics, computer-aided design, VLSI layout, and computer graphics. In a dynamic environment where objects are mobile, intersection or proximity among objects has obvious applications. Consider, for instance, the problem of collision detection in robot motion planning. The Euclidean distance is a commonly used measure in these areas. Numerous efficient algorithms are known..