Fusing sound and dead reckoning for multi-robot cooperative localization

Abstract

This paper presents a multi-robot distributed cooperative localization method based on the fusion of sound and dead reckoning. First, each robot computes its position and orientation by dead reckoning. In order to correct the accumulative error of the dead reckoning, we periodically fuse the dead-reckoning results with the sound based relative localization using the extended Kalman filter (EKF) algorithm. Two self-made small mobile robots are adopted to carry out indoor experiments, and unknown robot is equipped with cheap sound transmitting device and the known robot is equipped with sound receiving device. By comparing the dead-reckoning results, the EKF fusion results and the motion capture localization, we find that the proposed sound based EKF algorithm might be a new solution to the multi-robot localization problem.