Sound Source Localization experiment in variety real environment for intelligent service robots

In this paper, we present a Sound Source Localization (SSL) with robot auditory system for a network-based intelligent service robots that give a service in variety places. The main goal of this paper is to analysis performance of SSL algorithm in real environment for robot. We gathered the sound databases in museum, conference hall and restaurant for SSL experiments. These places will be served by intelligent robot in actually. With using these databases, we compared some of SSL algorithms and searched that we can use the SLL for intelligent robot in real environment. Algorithms for calculating time delay are based on the advanced cross correlation and GCC (Generalized Cross-Correlation)-PHAT(Phase Transform) by frequency characteristics for calculating time delay estimation. The robot platform used in this work is wever-R2, which is a network-based intelligent service robot developed at Intelligent Robot Research Division in ETRI.