Design and implementation of sound tracking multi-robot system in wireless sensor networks

A sound target-tracking multiple-robot system is described, including 4-channel microphone array for sound collection, magnetoresistive sensor for declination measurement, wireless senor network (WSN) for exchanging information. It has embedded sound signal enhancement, recognition and location method, and sound tracking strategy based on the digital signal processor (DSP). As the wireless network nodes, three robots comprise the WSN with personal compute (PC) in order to track the three different sound targets in task-oriented collaboration. Improved spectral subtraction method is used for noise reduction. As the feature of audio signal, Mel-frequency cepstral coefficient (MFCC) is extracted. Based on K-nearest neighbor classification method, we match the trained feature template to recognize sound signal type. Improved generalized cross correlation method is utilized to estimate time delay of arrival (TDOA), and then we employ spherical-interpolation for sound location according to the TDOA and the geometrical position of microphone array. A New mapping has been proposed to direct the motor to track sound target flexibly. As the sink node, PC receives and displays the processed result in WSN, and it also has the ultimate power to make decision on the received result in order to improve their accuracy.