Binaural Localization for a Mobile Sound Source

Abstract

This paper proposes a biologically inspired binaural auditory system to localize a mobile sound source. The robot considered in this paper has two microphones installed on a mobile platform to provide binaural hearing. An extended Kalman filter (EKF) is adopted to fuse the interaural time difference of arrival, which includes both the Doppler shift and the motion of the auditory system itself. The covariance matrix of EKF is utilized to determine a control policy based on optimizing the localization performance. Numerical simulations validate the proposed approach.