Ant colony optimization with parameter adaptation using fuzzy logic for TSP problems

Abstract

In this paper we propose a method for parameter adaptation in Ant Colony Optimization (ACO); with the use of a fuzzy system we dynamically adapt the “rho” parameter which is responsible for the evaporation of the pheromone trails in ACO. The main goal is to improve the results of ACO; basically the fuzzy system controls the ACO abilities for exploration and exploitation of the search space. The best problems to test ACO algorithms are the TSP problems, so a comparison with the proposed approach and others methods is performed and results discussed.