Design and Performance Evaluation of an Infotaxis-Based Three-Dimensional Algorithm for Odor Source Localization

Abstract

In this paper we tackle the problem of finding the source of a gaseous leak with a robot in a three-dimensional (3-D) physical space. The proposed method extends the operational range of the probabilistic Infotaxis algorithm [1] into 3-D and makes multiple improvements in order to increase its performance in such settings. The method has been tested systematically through high-fidelity simulations and in a wind tunnel emulating realistic conditions. The impact of multiple algorithmic and environmental parameters has been studied in the experiments. The algorithm shows good performance in various environmental conditions, particularly in high wind speeds and different source release rates.

Particle Source Localization With a Low-Cost Robotic Sensor System: Algorithmic Design and Performance Evaluation

“…Assuming P (C k |source = (x s , y s )) satisfies Poisson distribution (the same as [25], [48]), it can be calculated by…”

“…(3)) used in the algorithm, we first measure the baseline particle concentration in the clean air and then define multiple levels of concentrations for higher values. Similar to [25], [48], we set the robot to do a crosswind scan as its plume finding stage and record concentration values at the beginning of the experiments. The robot will come to the highest concentration point during this crosswind scan and start the algorithm.…”

“…We set the same fixed baseline B for all six modalities based on the measurement we get from the dominant particle size (which is the smallest size particle). Based on B, we define 10 concentration levels similar to the previous works [25], [48].…”

“…Only one of these conditions needs to be met to terminate the algorithm. The ending strategy is similar to our previous work [25], but we adjusted the related threshold according to the peculiarities of particle plumes.…”

“…Multiple slices of the obtained particle map perpendicular to the X-axis were used to calculate σ y and σ z . In a previous work fitting the chemical plume [25], both σ y and σ z were fitted into linear functions of x. In this paper, σ y was fitted into the empirical power law function [49] of x and σ z is fitted into the Nike function of x for decreasing error.…”