any monotone function is realized by interlocked polygons任何单调函数实现联锁多边形.pdfVIP

Algorithms 2012, 5, 148-157; doi:10.3390/a5010148 OPEN ACCESS algorithms ISSN 1999-4893 /journal/algorithms Article Any Monotone Function Is Realized by Interlocked Polygons Erik D. Demaine 1, Martin L. Demaine 1 and Ryuhei Uehara 2,* 1 Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, MA 02139, USA; E-Mails: edemaine@ (E.D.D.); mdemaine@ (M.L.D.) 2 School of Information Science, Japan Advanced Institute of Science and Technology, Ishikawa 923-1292, Japan ⋆ Author to whom correspondence should be addressed; E-Mail: uehara@jaist.ac.jp; Tel.: +81-761-51-1220; Fax: +81-761-51-1149. Received: 15 November 2011; in revised form: 13 March 2012 / Accepted: 14 March 2012 / Published: 19 March 2012 Abstract: Suppose there is a collection of simple polygons in the plane, none of which overlap each other. The polygons are interlocked if no subset can be separated arbitrarily far from the rest. It is natural to ask the characterization of the subsets that makes the set of interlocked polygons free (not interlocked). This abstracts the essence of a kind of sliding block puzzle. We show that any monotone Boolean function on variables can be described by interlocked polygons. We also show that the decision problem that asks if given polygons are interlocked is PSPACE-complete. Keywords: computational complexity; interlocked polygons; monotone boolean function; sliding block puzzle 1. Introduction Since Sam Loyd invented the famous 15-puzzle, sliding block puzzles have