Distributed Data Fusion for Multirobot Search

This is an author's peer-reviewed final manuscript, as accepted by the publisher. The published article is copyrighted by IEEE - Institute of Electrical and Electronics Engineers and can be found at: http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=70©2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.This paper presents novel data fusion methods\ud that enable teams of vehicles to perform target search tasks\ud without guaranteed communication. Techniques are introduced\ud for merging estimates of a target’s position from vehicles that\ud regain contact after long periods of time, and a fully distributed\ud team planning algorithm is proposed that utilizes limited shared\ud information as it becomes available. The proposed data fusion\ud techniques are shown to avoid overcounting information, which\ud ensures that combining data from different vehicles will not decrease\ud the performance of the search. Motivated by the underwater\ud search domain, a realistic underwater acoustic communication\ud channel is used to determine the probability of successful data\ud transfer between two locations. The channel model is integrated\ud into a simulation of multiple autonomous vehicles in both\ud open water and harbor environments. The results demonstrate\ud that the proposed distributed coordination techniques provide\ud performance competitive with full communication