Selección de un algoritmo de localización acústica para su uso en aplicaciones de robótica industrial

Abstract

Automation of vehicles requires a redundant network of sensors that can determine the position of the objects at all times. Localization of moving objects may be performed through the use of acoustic cues. Delay of Arrival (DOA) estimation is a popular method for determining the azimuth angle of a source with respect to a sensor array. Three separate algorithms were examined for a linear array of equidistant microphones, and evaluated based on performance, accuracy and computational resources consumed. The determination of the multichannel cross correlation coefficient for the array was found to be an efficient and accurate method for DOA estimation.