Journal of Robotics and Mechatronics
Online ISSN : 1883-8049
Print ISSN : 0915-3942
ISSN-L : 0915-3942
Special Issue on Robot Audition Technologies
Outdoor Acoustic Event Identification with DNN Using a Quadrotor-Embedded Microphone Array
Osamu SugiyamaSatoshi UemuraAkihide NagamineRyosuke KojimaKeisuke NakamuraKazuhiro Nakadai
Author information
JOURNAL OPEN ACCESS

2017 Volume 29 Issue 1 Pages 188-197

Details
Abstract

This paper addresses Acoustic Event Identification (AEI) of acoustic signals observed with a microphone array embedded in a quadrotor that is flying in a noisy outdoor environment. In such an environment, noise generated by rotors, wind, and other sound sources is a big problem. To solve this, we propose the use of a combination of two approaches that have recently been introduced: Sound Source Separation (SSS) and Sound Source Identification (SSI). SSS improves the Signal-to-Noise Ratio (SNR) of the input sound, and SSI is then performed on the SNR-improved sound. Two SSS methods are investigated. One is a single channel algorithm, Robust Principal Component Analysis (RPCA), and the other is Geometric High-order Decorrelation-based Source Separation (GHDSS-AS), known as a multichannel method. For SSI, we investigate two types of deep neural networks namely Stacked denoising Autoencoder (SdA) and Convolutional Neural Network (CNN), which have been extensively studied as highly-performant approaches in the fields of automatic speech recognition and visual object recognition. Preliminary experiments have showed the effectiveness of the proposed approaches, a combination of GHDSS-AS and CNN in particular. This combination correctly identified over 80% of sounds in an 8-class sound classification recorded by a hovering quadrotor. In addition, the CNN identifier that was implemented could be handled even with a low-end CPU by measuring the prediction time.

Content from these authors

This article cannot obtain the latest cited-by information.

© 2017 Fuji Technology Press Ltd.

This article is licensed under a Creative Commons [Attribution-NoDerivatives 4.0 International] license (https://creativecommons.org/licenses/by-nd/4.0/).
The journal is fully Open Access under Creative Commons licenses and all articles are free to access at JRM Official Site.
https://www.fujipress.jp/jrm/rb-about/
Previous article Next article
feedback
Top