A Genetic-Neuro Algorithm for Tiling Problems with Rotation and Reflection of Figures

Abstract

This paper describes an algorithm for tiling with polyminoes thatconsider rotation and/or reflection of figures in the steps of90degrees. First, we review the previous parallel algorithms for tilingproblems. Next, we propose a hybrid approach that is based on geneticalgorithms (GA)and artificial neural networks(ANN). In this approach,the production of new members in GA and their evaluation are performedby a Hopfield neural network. Finally we compare our method with theprevious works, and show that our method can produce global minima formany problems. The algorithm can be used for solving a variety of2D-packing problems.