Using Particle Filter and Machine Learning for Accuracy Estimation of Robot Localization

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

Abstract

Robot localization is a fundamental capability of all mobile robots. Because of uncertainties in acting and sensing and environmental factors such as people flocking around robots there is always the risk that a robot loses its localization. Very often behaviors of robots rely on a valid position estimation. Thus, for dependability of robot systems it is of great interest for the system to know the state of its localization component. In this paper we present an approach that allows a robot to asses if the localization is still valid. The approach assumes that the underlying localization approach is based on a particle filter. We use deep learning to identify temporal patterns in the particles in the case of losing/lost localization in combination with weak classifiers from the particle set and perception for boosted learning of a localization monitor. The approach is evaluated in a simulated transport robot environment where a degraded localization is provoked by disturbances cased by dynamic obstacles.
Originalspracheenglisch
TitelAdvances and Trends in Artificial Intelligence. From Theory to Practice
UntertitelIEA/AIE 2019
Redakteure/-innenFranz Wotawa, Gerhard Friedrich, Ingo Pill, Roxane Koitz-Hristov, Moonis Ali
Herausgeber (Verlag)Springer, Cham
Seiten700-713
Seitenumfang14
ISBN (elektronisch)978-3-030-22999-3
ISBN (Print)978-3-030-22998-6
DOIs
PublikationsstatusVeröffentlicht - 15 Jun 2019
Veranstaltung32nd International Conference on Industrial, Engineering & Other Applications of Applied Intelligent Systems - Graz, Österreich
Dauer: 9 Jul 201911 Jul 2019

Publikationsreihe

NameLecture Notes in Computer Science
Herausgeber (Verlag)Springer
Band11606

Konferenz

Konferenz32nd International Conference on Industrial, Engineering & Other Applications of Applied Intelligent Systems
KurztitelIEA/AIE 2019
LandÖsterreich
OrtGraz
Zeitraum9/07/1911/07/19

Fingerprint

Learning systems
Robots
Mobile robots
Classifiers

Dies zitieren

Eder, M. J., Reip, M., & Steinbauer, G. (2019). Using Particle Filter and Machine Learning for Accuracy Estimation of Robot Localization. in F. Wotawa, G. Friedrich, I. Pill, R. Koitz-Hristov, & M. Ali (Hrsg.), Advances and Trends in Artificial Intelligence. From Theory to Practice: IEA/AIE 2019 (S. 700-713). (Lecture Notes in Computer Science; Band 11606). Springer, Cham. https://doi.org/10.1007/978-3-030-22999-3_60

Using Particle Filter and Machine Learning for Accuracy Estimation of Robot Localization. / Eder, Matthias Josef; Reip, Michael; Steinbauer, Gerald.

Advances and Trends in Artificial Intelligence. From Theory to Practice: IEA/AIE 2019. Hrsg. / Franz Wotawa; Gerhard Friedrich; Ingo Pill; Roxane Koitz-Hristov; Moonis Ali. Springer, Cham, 2019. S. 700-713 (Lecture Notes in Computer Science; Band 11606).

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

Eder, MJ, Reip, M & Steinbauer, G 2019, Using Particle Filter and Machine Learning for Accuracy Estimation of Robot Localization. in F Wotawa, G Friedrich, I Pill, R Koitz-Hristov & M Ali (Hrsg.), Advances and Trends in Artificial Intelligence. From Theory to Practice: IEA/AIE 2019. Lecture Notes in Computer Science, Bd. 11606, Springer, Cham, S. 700-713, Graz, Österreich, 9/07/19. https://doi.org/10.1007/978-3-030-22999-3_60
Eder MJ, Reip M, Steinbauer G. Using Particle Filter and Machine Learning for Accuracy Estimation of Robot Localization. in Wotawa F, Friedrich G, Pill I, Koitz-Hristov R, Ali M, Hrsg., Advances and Trends in Artificial Intelligence. From Theory to Practice: IEA/AIE 2019. Springer, Cham. 2019. S. 700-713. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-030-22999-3_60
Eder, Matthias Josef ; Reip, Michael ; Steinbauer, Gerald. / Using Particle Filter and Machine Learning for Accuracy Estimation of Robot Localization. Advances and Trends in Artificial Intelligence. From Theory to Practice: IEA/AIE 2019. Hrsg. / Franz Wotawa ; Gerhard Friedrich ; Ingo Pill ; Roxane Koitz-Hristov ; Moonis Ali. Springer, Cham, 2019. S. 700-713 (Lecture Notes in Computer Science).
@inproceedings{ec4474e1d75f4852a27f2ed32e452e28,
title = "Using Particle Filter and Machine Learning for Accuracy Estimation of Robot Localization",
abstract = "Robot localization is a fundamental capability of all mobile robots. Because of uncertainties in acting and sensing and environmental factors such as people flocking around robots there is always the risk that a robot loses its localization. Very often behaviors of robots rely on a valid position estimation. Thus, for dependability of robot systems it is of great interest for the system to know the state of its localization component. In this paper we present an approach that allows a robot to asses if the localization is still valid. The approach assumes that the underlying localization approach is based on a particle filter. We use deep learning to identify temporal patterns in the particles in the case of losing/lost localization in combination with weak classifiers from the particle set and perception for boosted learning of a localization monitor. The approach is evaluated in a simulated transport robot environment where a degraded localization is provoked by disturbances cased by dynamic obstacles.",
author = "Eder, {Matthias Josef} and Michael Reip and Gerald Steinbauer",
year = "2019",
month = "6",
day = "15",
doi = "10.1007/978-3-030-22999-3_60",
language = "English",
isbn = "978-3-030-22998-6",
series = "Lecture Notes in Computer Science",
publisher = "Springer, Cham",
pages = "700--713",
editor = "Franz Wotawa and Gerhard Friedrich and Ingo Pill and Roxane Koitz-Hristov and Moonis Ali",
booktitle = "Advances and Trends in Artificial Intelligence. From Theory to Practice",

}

TY - GEN

T1 - Using Particle Filter and Machine Learning for Accuracy Estimation of Robot Localization

AU - Eder, Matthias Josef

AU - Reip, Michael

AU - Steinbauer, Gerald

PY - 2019/6/15

Y1 - 2019/6/15

N2 - Robot localization is a fundamental capability of all mobile robots. Because of uncertainties in acting and sensing and environmental factors such as people flocking around robots there is always the risk that a robot loses its localization. Very often behaviors of robots rely on a valid position estimation. Thus, for dependability of robot systems it is of great interest for the system to know the state of its localization component. In this paper we present an approach that allows a robot to asses if the localization is still valid. The approach assumes that the underlying localization approach is based on a particle filter. We use deep learning to identify temporal patterns in the particles in the case of losing/lost localization in combination with weak classifiers from the particle set and perception for boosted learning of a localization monitor. The approach is evaluated in a simulated transport robot environment where a degraded localization is provoked by disturbances cased by dynamic obstacles.

AB - Robot localization is a fundamental capability of all mobile robots. Because of uncertainties in acting and sensing and environmental factors such as people flocking around robots there is always the risk that a robot loses its localization. Very often behaviors of robots rely on a valid position estimation. Thus, for dependability of robot systems it is of great interest for the system to know the state of its localization component. In this paper we present an approach that allows a robot to asses if the localization is still valid. The approach assumes that the underlying localization approach is based on a particle filter. We use deep learning to identify temporal patterns in the particles in the case of losing/lost localization in combination with weak classifiers from the particle set and perception for boosted learning of a localization monitor. The approach is evaluated in a simulated transport robot environment where a degraded localization is provoked by disturbances cased by dynamic obstacles.

U2 - 10.1007/978-3-030-22999-3_60

DO - 10.1007/978-3-030-22999-3_60

M3 - Conference contribution

SN - 978-3-030-22998-6

T3 - Lecture Notes in Computer Science

SP - 700

EP - 713

BT - Advances and Trends in Artificial Intelligence. From Theory to Practice

A2 - Wotawa, Franz

A2 - Friedrich, Gerhard

A2 - Pill, Ingo

A2 - Koitz-Hristov, Roxane

A2 - Ali, Moonis

PB - Springer, Cham

ER -