New Strategies Based on Improved Fruit Fly Optimization Algorithm for Unknown Indoor Odor Source Location

Abstract

It has great importance to locate the leaked source of hazardous odor in a ventilated indoor environment, for chemical warehouse safety and anti-terrorist explosion, etc. The distribution of odor plume is time-varying and intermittent in the indoor environment, which makes it difficult for single robot to locate the odor source. The multi-robot cooperative location of odor sources can reduce the interference of the above features and achieve high success rate of odor source localization. In this paper, the fruit fly optimization algorithm is applied to the odor source location problem, and a Fixed-step Fruit Fly Optimization Algorithm (FS-FOA) is proposed. To further improve the efficiency of the search and localization in the larger search range, a new strategy with adaptive algorithm for odor source localization is put forward, which is Concentration-Adaptive -Step Fruit Fly Optimization Algorithm (CAS-FOA). And the efficiency of the strategy in dynamic plume environment is validated in our experiments.