The ant colony metaphor in continuous spaces using boundary search

This paper presents an application of the ant colony metaphor for continuous space optimization problems. The ant algortihm proposed works following the principle of the ant colony approach, i.e., a population of agents iteratively, cooperatively, and independently search for a solution. Each ant in the distributed algorithm applies a local search operator which explores the neighborhood region of a particular point in the search space (individual search level). The local search operator is designed for exploring the boundary between the feasible and infeasible search space. On the other hand, each ant obtains global information from the colony in order to exploit the more promising regions of the search space (cooperation level). The ant colony based algorithm presented here was successfully applied to two widely studied and interesting constrained numerical optimization test cases. Eje: Agentes y Sistemas Inteligentes (ASI) Red de Universidades con Carreras en Informática (RedUNCI)