Active Localization of Gas Leaks using Fluid Simulation

Martin Asenov, Marius Rutkauskas, Derryck Reid, Kartic Subr, Subramanian Ramamoorthy

Research output: Contribution to journalArticle

Abstract

Sensors are routinely mounted on robots to acquire various forms of measurements in spatiotemporal fields. Locating features within these fields and reconstruction (mapping) of the dense fields can be challenging in resource-constrained situations, such as when trying to locate the source of a gas leak from a small number of measurements. In such cases, a model of the underlying complex dynamics can be exploited to discover informative paths within the field. We use a fluid simulator as a model to guide inference for the location of a gas leak. We perform localization via minimization of the discrepancy between observed measurements and gas concentrations predicted by the simulator. Our method is able to account for dynamically varying parameters of wind flow (e.g., direction and strength) and its effects on the observed distribution of gas. We develop algorithms for offline inference as well as for online path discovery via active sensing. We demonstrate the efficiency, accuracy, and versatility of our algorithm using experiments with a physical robot conducted in outdoor environments. We deploy an unmanned air vehicle mounted with a CO $_2$ sensor to automatically seek out a gas cylinder emitting CO $_2$ via a nozzle. We evaluate the accuracy of our algorithm by measuring the error in the inferred location of the nozzle, based on which we show that our proposed approach is competitive with respect to state-of-the-art baselines.
LanguageEnglish
Pages1776-1783
Number of pages8
JournalIEEE Robotics and Automation Letters
Volume4
Issue number2
Early online date29 Jan 2019
DOIs
Publication statusPublished - Apr 2019

Fingerprint

Fluids
Gases
Nozzles
Simulators
Robots
Gas cylinders
Sensors
Air
Experiments

Cite this

Asenov, Martin ; Rutkauskas, Marius ; Reid, Derryck ; Subr, Kartic ; Ramamoorthy, Subramanian. / Active Localization of Gas Leaks using Fluid Simulation. In: IEEE Robotics and Automation Letters. 2019 ; Vol. 4, No. 2. pp. 1776-1783.
@article{d2e368e64e794af09bc9ea85aecb6369,
title = "Active Localization of Gas Leaks using Fluid Simulation",
abstract = "Sensors are routinely mounted on robots to acquire various forms of measurements in spatiotemporal fields. Locating features within these fields and reconstruction (mapping) of the dense fields can be challenging in resource-constrained situations, such as when trying to locate the source of a gas leak from a small number of measurements. In such cases, a model of the underlying complex dynamics can be exploited to discover informative paths within the field. We use a fluid simulator as a model to guide inference for the location of a gas leak. We perform localization via minimization of the discrepancy between observed measurements and gas concentrations predicted by the simulator. Our method is able to account for dynamically varying parameters of wind flow (e.g., direction and strength) and its effects on the observed distribution of gas. We develop algorithms for offline inference as well as for online path discovery via active sensing. We demonstrate the efficiency, accuracy, and versatility of our algorithm using experiments with a physical robot conducted in outdoor environments. We deploy an unmanned air vehicle mounted with a CO $_2$ sensor to automatically seek out a gas cylinder emitting CO $_2$ via a nozzle. We evaluate the accuracy of our algorithm by measuring the error in the inferred location of the nozzle, based on which we show that our proposed approach is competitive with respect to state-of-the-art baselines.",
author = "Martin Asenov and Marius Rutkauskas and Derryck Reid and Kartic Subr and Subramanian Ramamoorthy",
year = "2019",
month = "4",
doi = "10.1109/LRA.2019.2895820",
language = "English",
volume = "4",
pages = "1776--1783",
journal = "IEEE Robotics and Automation Letters",
issn = "2377-3766",
publisher = "IEEE",
number = "2",

}

Active Localization of Gas Leaks using Fluid Simulation. / Asenov, Martin; Rutkauskas, Marius; Reid, Derryck; Subr, Kartic; Ramamoorthy, Subramanian.

In: IEEE Robotics and Automation Letters, Vol. 4, No. 2, 04.2019, p. 1776-1783.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Active Localization of Gas Leaks using Fluid Simulation

AU - Asenov, Martin

AU - Rutkauskas, Marius

AU - Reid, Derryck

AU - Subr, Kartic

AU - Ramamoorthy, Subramanian

PY - 2019/4

Y1 - 2019/4

N2 - Sensors are routinely mounted on robots to acquire various forms of measurements in spatiotemporal fields. Locating features within these fields and reconstruction (mapping) of the dense fields can be challenging in resource-constrained situations, such as when trying to locate the source of a gas leak from a small number of measurements. In such cases, a model of the underlying complex dynamics can be exploited to discover informative paths within the field. We use a fluid simulator as a model to guide inference for the location of a gas leak. We perform localization via minimization of the discrepancy between observed measurements and gas concentrations predicted by the simulator. Our method is able to account for dynamically varying parameters of wind flow (e.g., direction and strength) and its effects on the observed distribution of gas. We develop algorithms for offline inference as well as for online path discovery via active sensing. We demonstrate the efficiency, accuracy, and versatility of our algorithm using experiments with a physical robot conducted in outdoor environments. We deploy an unmanned air vehicle mounted with a CO $_2$ sensor to automatically seek out a gas cylinder emitting CO $_2$ via a nozzle. We evaluate the accuracy of our algorithm by measuring the error in the inferred location of the nozzle, based on which we show that our proposed approach is competitive with respect to state-of-the-art baselines.

AB - Sensors are routinely mounted on robots to acquire various forms of measurements in spatiotemporal fields. Locating features within these fields and reconstruction (mapping) of the dense fields can be challenging in resource-constrained situations, such as when trying to locate the source of a gas leak from a small number of measurements. In such cases, a model of the underlying complex dynamics can be exploited to discover informative paths within the field. We use a fluid simulator as a model to guide inference for the location of a gas leak. We perform localization via minimization of the discrepancy between observed measurements and gas concentrations predicted by the simulator. Our method is able to account for dynamically varying parameters of wind flow (e.g., direction and strength) and its effects on the observed distribution of gas. We develop algorithms for offline inference as well as for online path discovery via active sensing. We demonstrate the efficiency, accuracy, and versatility of our algorithm using experiments with a physical robot conducted in outdoor environments. We deploy an unmanned air vehicle mounted with a CO $_2$ sensor to automatically seek out a gas cylinder emitting CO $_2$ via a nozzle. We evaluate the accuracy of our algorithm by measuring the error in the inferred location of the nozzle, based on which we show that our proposed approach is competitive with respect to state-of-the-art baselines.

U2 - 10.1109/LRA.2019.2895820

DO - 10.1109/LRA.2019.2895820

M3 - Article

VL - 4

SP - 1776

EP - 1783

JO - IEEE Robotics and Automation Letters

T2 - IEEE Robotics and Automation Letters

JF - IEEE Robotics and Automation Letters

SN - 2377-3766

IS - 2

ER -