Expanding the Knowledge Horizon in Underwater Robot Swarms

• Published in: arXiv.org
• Main Authors: Fioriti, Enzo, Chiesa, Stefano, Fratichini, Fabio
• Format: Paper
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 19.08.2013
• Subjects: Gaussian distribution; Nodes; Normal distribution; Statistical analysis; Topology; Underwater robots
• ISSN: 2331-8422

In this paper we study the time delays affecting the diffusion of information in an underwater heterogeneous robot swarm, considering a time-sensitive environment. In many situations each member of the swarm must update its knowledge about the environment as soon as possible, thus every effort to expand the knowledge horizon is useful. Otherwise critical information may not reach nodes far from the source causing dangerous misbehaviour of the swarm. We consider two extreme situations. In the first scenario we have an unique probabilistic delay distribution. In the second scenario, each agent is subject to a different truncated gaussian distribution, meaning local conditions are significantly different from link to link. We study how several swarm topologies react to the two scenarios and how to allocate the more efficient transmission resources in order to expand the horizon. Results show that significant time savings under a gossip-like protocol are possible properly allocating the resources. Moreover, methods to determine the fastest swarm topologies and the most important nodes are suggested.