Connectivity vs. control: Using directional and positional cues to stabilize routing in robot networks

Various coordination algorithms have been proposed for robot networks. One of the fundamental assumptions of such algorithms is that the underlying connectivity graph be stable. Adhoc routing protocols attempt to optimize the path from source to destination and do not guarantee route stability. We bridge this gap by providing directional and locational cues to the routing protocol to provide more stable routes. We implement our ideas on optimized link state routing (OLSR), a popular proactive routing protocol for robot networks. Our results show that simple directional and locational cues can achieve up to 20% fewer route switches in comparison to the basic version of OLSR.