Statistical Evaluation of the Kernel DM+V/W Algorithm for Building Gas Distribution Maps in Uncontrolled Environments

Matteo Reggente; Achim J. Lilienthal

doi:10.1016/j.proche.2009.07.120

Statistical Evaluation of the Kernel DM+V/W Algorithm for Building Gas Distribution Maps in Uncontrolled Environments

Matteo Reggente, Achim J. Lilienthal

Örebro University

Research output: Contribution to journal › Conference article › peer-review

6 Scopus citations

Abstract

In this paper we present a statistical evaluation of the Kernel DM+V/W algorithm to build two-dimensional gas distribution maps with a mobile robot. In addition to gas sensor measurements from an "e-nose" the Kernel DM+V/W algorithm also takes into account wind information received from an ultrasonic anemometer. We evaluate the method based on real measurements in three uncontrolled environments with very different properties. As a measure for the model quality we compute how well unseen measurements are predicted in terms of the data likelihood. A paired Wilcoxon signed rank test shows a significant improvement (at a confidence level of 95%) of the model quality when using wind information.

Original language	English
Pages (from-to)	481-484
Number of pages	4
Journal	Procedia Chemistry
Volume	1
Issue number	1
DOIs	https://doi.org/10.1016/j.proche.2009.07.120
State	Published - Sep 2009
Externally published	Yes
Event	Eurosensors 23rd Conference - Lausanne, Switzerland Duration: 6 Sep 2009 → 9 Sep 2009

Keywords

e-nose
gas distribution
gas sensing
kernel density estimation
mobile robots
model evaluation

Access to Document

10.1016/j.proche.2009.07.120

Cite this

@article{ef435ae099534f2f97d67372ff7a0fdc,

title = "Statistical Evaluation of the Kernel DM+V/W Algorithm for Building Gas Distribution Maps in Uncontrolled Environments",

abstract = "In this paper we present a statistical evaluation of the Kernel DM+V/W algorithm to build two-dimensional gas distribution maps with a mobile robot. In addition to gas sensor measurements from an {"}e-nose{"} the Kernel DM+V/W algorithm also takes into account wind information received from an ultrasonic anemometer. We evaluate the method based on real measurements in three uncontrolled environments with very different properties. As a measure for the model quality we compute how well unseen measurements are predicted in terms of the data likelihood. A paired Wilcoxon signed rank test shows a significant improvement (at a confidence level of 95%) of the model quality when using wind information.",

keywords = "e-nose, gas distribution, gas sensing, kernel density estimation, mobile robots, model evaluation",

author = "Matteo Reggente and Lilienthal, {Achim J.}",

year = "2009",

month = sep,

doi = "10.1016/j.proche.2009.07.120",

language = "English",

volume = "1",

pages = "481--484",

journal = "Procedia Chemistry",

issn = "1876-6196",

publisher = "Elsevier B.V.",

number = "1",

note = "Eurosensors 23rd Conference ; Conference date: 06-09-2009 Through 09-09-2009",

}

TY - JOUR

T1 - Statistical Evaluation of the Kernel DM+V/W Algorithm for Building Gas Distribution Maps in Uncontrolled Environments

AU - Reggente, Matteo

AU - Lilienthal, Achim J.

PY - 2009/9

Y1 - 2009/9

N2 - In this paper we present a statistical evaluation of the Kernel DM+V/W algorithm to build two-dimensional gas distribution maps with a mobile robot. In addition to gas sensor measurements from an "e-nose" the Kernel DM+V/W algorithm also takes into account wind information received from an ultrasonic anemometer. We evaluate the method based on real measurements in three uncontrolled environments with very different properties. As a measure for the model quality we compute how well unseen measurements are predicted in terms of the data likelihood. A paired Wilcoxon signed rank test shows a significant improvement (at a confidence level of 95%) of the model quality when using wind information.

AB - In this paper we present a statistical evaluation of the Kernel DM+V/W algorithm to build two-dimensional gas distribution maps with a mobile robot. In addition to gas sensor measurements from an "e-nose" the Kernel DM+V/W algorithm also takes into account wind information received from an ultrasonic anemometer. We evaluate the method based on real measurements in three uncontrolled environments with very different properties. As a measure for the model quality we compute how well unseen measurements are predicted in terms of the data likelihood. A paired Wilcoxon signed rank test shows a significant improvement (at a confidence level of 95%) of the model quality when using wind information.

KW - e-nose

KW - gas distribution

KW - gas sensing

KW - kernel density estimation

KW - mobile robots

KW - model evaluation

UR - http://www.scopus.com/inward/record.url?scp=71649108815&partnerID=8YFLogxK

U2 - 10.1016/j.proche.2009.07.120

DO - 10.1016/j.proche.2009.07.120

M3 - Conference article

AN - SCOPUS:71649108815

SN - 1876-6196

VL - 1

SP - 481

EP - 484

JO - Procedia Chemistry

JF - Procedia Chemistry

IS - 1

T2 - Eurosensors 23rd Conference

Y2 - 6 September 2009 through 9 September 2009

ER -

Statistical Evaluation of the Kernel DM+V/W Algorithm for Building Gas Distribution Maps in Uncontrolled Environments

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this