Authors: Antonio Gonzalez-Pardo; Javier Del Ser; David Camacho
Addresses: Basque Center for Applied Mathematics (BCAM), 48009, Bilbao, Basque Country, Spain; OPTIMA Area, TECNALIA, 48160 Derio, Spain ' OPTIMA Area, TECNALIA, 48160 Derio, Spain ' Universidad Autónoma de Madrid, 28049, Madrid, Spain
Abstract: In the last years, there have been a huge increase in the number of research contributions that use games and video-games as an application domain for testing different artificial intelligence algorithms. Some of these problems can be represented as a constraint satisfaction problem (CSP), and heuristics algorithms (such as ant colony optimisation) can be used due to the complexity of the modelled problems. This paper presents a comparative study of the performance of a novel ACO model for CSP-based board games. In this work, two different oblivion rate meta-heuristics for controlling the number of pheromones created in the model have been created. Experimental results reveal that both meta-heuristics reduce considerably the number of pheromones produced in the system without affecting the quality of the solutions in terms of average optimality.
Keywords: ant colony optimisation; pheromone control; strategy board games; N-Queens problem; oblivion rate.
International Journal of Bio-Inspired Computation, 2017 Vol.10 No.2, pp.136 - 144
Received: 05 Dec 2014
Accepted: 04 Jun 2015
Published online: 28 Jul 2017 *