elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Material Classification through Knocking and Grasping by Learning of Structure-Borne Sound under Changing Acoustic Conditions

Neumann, Michael and Nottensteiner, Korbinian and Kossyk, Ingo and Marton, Zoltan-Csaba (2018) Material Classification through Knocking and Grasping by Learning of Structure-Borne Sound under Changing Acoustic Conditions. In: IEEE International Conference on Automation Science and Engineering. 14th IEEE International Conference on Automation Science and Engineering (CASE 2018), 2018-08-20 - 2018-08-24, Munich. doi: 10.1109/COASE.2018.8560527. ISBN 978-1-5386-3593-3. ISSN 2161-8089.

[img] PDF - Only accessible within DLR
4MB

Official URL: https://ieeexplore.ieee.org/document/8560527

Abstract

Structure-borne sound is an interesting sensory modality for inferring contact information in robotics due to comparably cheaply available sensor hardware, the possibility to integrate it into existing systems with little effort, and due to the richness of information in the acquired signals. In this work we investigate whether its is feasible to fit a robotic system with piezo acoustics sensorics in order to infer on properties about objects in the workspace of the robot during contact events. In contrast to existing works regarding object and material identification by evaluating sound the challenge in our experimental setup is that the sensor is integrated in the structure of the robot, hence, the measured audio signal are not only governed by the acoustic properties of the objects we try to identify but are also strongly influenced by the changing resonance properties of the robot due to its kinematic configuration and the ego-noise during operation. Therefore, we investigate whether it is possible to learn a classifier that is invariant and robust to these configuration dependent changes in the acquired audio signals. We exemplarily show the feasibility of this approach for contact inference in a material classification experiment and compare the performance of a deep learning classifier to several baseline machine learning methods. We found that a representation learning approach using a deep neural network shows the highest invariance to the changing acoustics properties and outperforms the baseline methods in our experiments. The results encourage the further investigation of the uses of structure-borne sound in robotic applications.

Item URL in elib:https://elib.dlr.de/124135/
Document Type:Conference or Workshop Item (Speech)
Title:Material Classification through Knocking and Grasping by Learning of Structure-Borne Sound under Changing Acoustic Conditions
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Neumann, MichaelUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Nottensteiner, KorbinianUNSPECIFIEDhttps://orcid.org/0000-0002-6016-6235UNSPECIFIED
Kossyk, IngoUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Marton, Zoltan-CsabaUNSPECIFIEDhttps://orcid.org/0000-0002-3035-493XUNSPECIFIED
Date:2018
Journal or Publication Title:IEEE International Conference on Automation Science and Engineering
Refereed publication:Yes
Open Access:No
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:No
DOI:10.1109/COASE.2018.8560527
ISSN:2161-8089
ISBN:978-1-5386-3593-3
Status:Published
Keywords:material classification; deep learning; structure-borne sound; knocking; grasping; perception; robotic systems; variational auto-encoder
Event Title:14th IEEE International Conference on Automation Science and Engineering (CASE 2018)
Event Location:Munich
Event Type:international Conference
Event Start Date:20 August 2018
Event End Date:24 August 2018
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Space System Technology
DLR - Research area:Raumfahrt
DLR - Program:R SY - Space System Technology
DLR - Research theme (Project):R - Vorhaben Intelligente Mobilität (old)
Location: Oberpfaffenhofen
Institutes and Institutions:Institute of Robotics and Mechatronics (since 2013)
Deposited By: Nottensteiner, Korbinian
Deposited On:03 Dec 2018 16:33
Last Modified:24 Apr 2024 20:28

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.