Naslov
Applying von Mises Distribution to Microphone Array Probabilistic Sensor Modelling

Autori
Marković, Ivan ; Petrović, Ivan

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings for the Joint Conference of ISR 2010 (41st Internationel Symposium on Robotics) und ROBOTIK 2010 (6th German Conference on Robotics) / - München, 2010, 21-27

Skup
The joint conference of the 41st International Symposium on Robotics (ISR 2010) and the 6th German Conference on Robotics (ROBOTIK 2010)

Mjesto i datum
München, Njemačka, 07.07.2010. - 09.07.2010

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Speaker localization; microphone array; von Mises distribution; Particle Filtering

Sažetak
This paper deals with the problem of localizing and tracking a moving speaker over the full range around the mobile robot. The proposed algorithm is based on estimating the time difference of arrival by maximizing the weighted cross-correlation function in order to, by series of geometrical calculation, determine the azimuth angle of the detected speaker. A post processing technique is proposed in which each of these microphone-pair determined azimuths are further combined into a mixture of the von Mises distributions, thus producing a practical probabilistic representation of the microphone array measurement. It is shown that this distribution is inherently multimodal and that the system at hand is non-linear, which required a discrete representation of the distribution function by means of particle filtering. Experiments were conducted and the results show that the algorithm can reliably and accurately localize and track a moving sound source.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Temeljne tehničke znanosti

POVEZANOST RADA