An Improved On-line Strategy for Exploring Unknown Polygons

We present a new, on-line strategy for a mobile robot to explore an unknown simple polygon P, so as to output a so-called watchman route such that every interior point of P is visible from at least one point along the route. The length of the robot's route is guaranteed to be at most 6.7 times that of the shortest watchman route that could be computed off-line. This significantly improves upon the previously known 26.5-competitive strategy. A novelty of our strategy is an on-line implementation of a previously known off-line algorithm that approximates the optimum watchman route to a factor of $$\sqrt{2}$$2. The other is in the way the polygon exploration problem is decomposed into two different types of the subproblems and a new method for analyzing its cost performance.

[1]  Tomio Hirata,et al.  An incremental algorithm for constructing shortest watchman routes , 1993, Int. J. Comput. Geom. Appl..

[2]  Rolf Klein,et al.  The Polygon Exploration Problem , 2001, SIAM J. Comput..

[3]  Xiaotie Deng,et al.  How to learn an unknown environment , 1991, [1991] Proceedings 32nd Annual Symposium of Foundations of Computer Science.

[4]  Simeon C. Ntafos,et al.  Optimum Watchman Routes , 1988, Inf. Process. Lett..

[5]  Xuehou Tan,et al.  Approximation algorithms for the watchman route and zookeeper's problems , 2001, Discret. Appl. Math..

[6]  Moshe Dror,et al.  Touring a sequence of polygons , 2003, STOC '03.

[7]  XUEHOU TAN,et al.  Corrigendum to "An Incremental Algorithm for Constructing Shortest Watchman Routes" , 1999, Int. J. Comput. Geom. Appl..