Autonomous Navigation of Wireless Robot Swarms with Covert Leaders

Abstract

The integration of advanced computation, wireless communication, and control technologies has facilitated the creation of autonomous robot swarms for many civil and military applications. In nature, animals that travel in groups often rely on social interactions among group members to make movement decisions. In many cases, few individuals within the group have pertinent knowledge about the destination and/or migration routes. In this paper, we adapt a swarm model developed for animal groups to study the unique problems associated with covert leadership in the context of wireless robot swarms. We term this problem autonomous navigation with covert leaders. In this covert leadership problem, only a small subset of robots in a robot swarm possess extra information that guides their movement, and both this information and the identities of those individuals possessing this information remain covert (to minimize the chance of being compromised). We describe a distributed navigation algorithm, where each robot locally makes its movement decision solely based on one-hop information collected via wireless communications. The effectiveness and merits of the described navigation algorithm are demonstrated through extensive simulations.