Explorative Particle Swarm Optimization Method for Gas/Odor Source Localization in an Indoor Environment with no Strong Airflow

摘要

This paper presents an algorithm to localize a gas source by using a swarm of robots in a large indoor environment without the presence of a relevant wind. The algorithm is composed of two different phases: an exploration phase aiming at finding a clue of the presence of a gas source and a localization phase to detect the emitting source. We propose a modified version of the particle swarm optimization (PSO) algorithm, called explorative particle swarm optimization (EPSO), as a strategy of movement for the swarm of robots during the localization phase. The introduced modifications point at avoiding multiple gas samplings in nearby locations, in this way increasing the exploration of the area and limiting the possibility for the swarm of being trapped in local maxima. Results from computer simulations are reported and discussed. They show that, in these environmental conditions, the proposed algorithm improves the performance of the standard PSO and can be a viable solution in localizing a chemical source.