Mobile robots in real-life settings would benefit from being able to localizesound sources. Such a capability can nicely complement vision to help localizea person or an interesting event in the environment, and also to provideenhanced processing for other capabilities such as speech recognition. In thispaper we present a robust sound source localization method in three-dimensionalspace using an array of 8 microphones. The method is based on afrequency-domain implementation of a steered beamformer along with aprobabilistic post-processor. Results show that a mobile robot can localize inreal time multiple moving sources of different types over a range of 5 meterswith a response time of 200 ms.
展开▼