Multi-robot smell source positioning method based on fruit fly optimization algorithm

Abstract

The invention relates to a multi-robot smell source positioning method based on a fruit fly optimization algorithm. The method is characterized in that the method comprises following steps of 1), smoke plume discovering: randomly presetting a robot group initial position Pos, dispersing robots in a diamond region with adaptive step length Step*2 by taking the initial position as the center, and determining whether to enter a smoke plume tracking stage according to the fact whether the obtained biggest concentration value is larger than a set concentration threshold value alpha; 2) smoke plume tracking: selecting a robot with the biggest smell concentration as an optimal individual Bestsmell, and if the concentration of the individual is larger than the group optimum Smellbest, updating the position to be the group optimum; and 3) smell source confirmation: using a set concentration threshold value C<lim> as an iteration termination condition, and when the optimal solution continuously found for 3 times meets a termination condition within the biggest iteration times, confirming that the task of positioning the smell source is finished. According to the invention, the fixed step length in the original fruit fly algorithm is corrected, adaptive change is performed according to the concentration and smell source positioning efficiency is effectively improved.

Description

technical field [0001] The invention relates to the research of a multi-robot odor source positioning method based on a fruit fly optimization algorithm, and belongs to the field of active olfaction of robots. Background technique [0002] Due to the frequent occurrence of damage to life and property due to the leakage of dangerous gases, many creatures in nature can judge the existence of danger and avoid danger in time by virtue of their sensitive sense of smell. Inspired by the effective use of olfactory information by organisms, some scholars have applied robots equipped with gas sensors to the research of locating dangerous chemical sources, and defined such research as robotic active olfaction. In the field of active olfactory research of robots, the source of dangerous gas leakage is defined as the odor source. [0003] Hayes defined the general odor source localization problem as the process of finding a single odor source as quickly as possible in a two-dimensional...

AI Technical Summary

Problems solved by technology

At present, the scope of application of the fruit fly optimization algorithm is mostly in the field of parameter optimization of continuous functions, and there is no direct application to the optimization problem in the discrete domain.

Images