Polygon Partitions

In 1973, Victor Klee posed the problem of determining the minimum number of guards sufficient to cover the interior of an n-wall art gallery room (Honsberger 1976). He posed this question extemporaneously in response to a request from Vasek Chvatal (at a conference at Stanford in August) for an interesting geometric problem, and Chvatal soon established what has become known as "Chvatal's Art Gallery Theorem" (or sometimes, "watchman theorem"): [n/3\ guards are occasionally necessary and always sufficient to cover a polygon of n vertices (Chvatal 1975). This simple and beautiful theorem has since been extended by mathematicians in several directions, and has been further developed by computer scientists studying partitioning algorithms. Now, a little more than a decade after Klee posed his question, there are enough related results to fill a book. By no means do all these results flow directly from Klee's problem, but there is a cohesion in the material presented here that is consistent with the spirit of his question. This chapter examines the original art gallery theorem and its associated algorithm. The algorithm leads to a discussion of triangulation, and a reexamination of the problem brings us to convex partitioning. The common theme throughout the chapter is polygon partitioning. Subsequent chapters branch off into specializations and generalizations of the original art gallery theorem and related algorithmic issues.