Intelligent sound source localization for dynamic environments

Authors:
Keisuke Nakamura;Kazuhiro Nakadai;Futoshi Asano;Yuji Hasegawa;Hiroshi Tsujino
Affiliations:
Department of Mechenical and Control Engineering, Tokyo Institute of Technology, Tokyo, Japan;Honda Research Institute Japan Co., Ltd., Wako, Saitama, Japan and Department of Information Science and Engineering, Tokyo Institute of Technology;Information Technology Research Institute, National Institute of Advanced Industrial Science and Technology, Tsukuba Central 2, Tsukuba, Japan and Honda Research Institute Japan Co., Ltd.;Honda Research Institute Japan Co., Ltd.;Honda Research Institute Japan Co., Ltd.
Venue:
IROS'09 Proceedings of the 2009 IEEE/RSJ international conference on Intelligent robots and systems
Year:
2009

Citing 4
Cited 0

Array Signal Processing: Concepts and Techniques

Array Signal Processing: Concepts and Techniques
Active Audition for Humanoid

Proceedings of the Seventeenth National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications of Artificial Intelligence
Computational Auditory Scene Analysis and Its Application to Robot Audition: Five Years Experience

ICKS '07 Proceedings of the Second International Conference on Informatics Research for Development of Knowledge Society Infrastructure
Broadband nearfield beamforming using a radial beampatterntransformation

IEEE Transactions on Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

As robotic technology plays an increasing role in human lives, "robot audition", human-robot communication, is of great interest, and robot audition needs to be robust and adaptable for dynamic environments. This paper addresses sound source localization working in dynamic environments for robots. Previously, noise robustness and dynamic localized sound selection have been enormous issues for practical use. To correct the issues, a new localization system "Selective Attention System" is proposed. The system has four new functions: localization with Generalized EigenValue Decomposition of correlation matrices for noise robustness ("Localization with GEVD"), sound source cancellation and focus ("Target Source Selection"), human-like dynamic Focus of Attention ("Dynamic FoA"), and correlation matrix estimation for robotic head rotation ("Correlation Matrix Estimation"). All are achieved by the dynamic design of correlation matrices. The system is implemented into a humanoid robot, and the experimental validation is successfully verified even when the robot microphones move dynamically.