Supporting Geographically Widespread UAV Swarms Through Graph-Based Network Relaying

Due to the increasing popularity of UAVs, we can observe a growing interest in multi-UAV applications. Many applications such as precision agriculture, border surveillance, mobile infrastructure for the IoT, etc., can benefit greatly if the UAVs are working together. However, to enable this cooperation, the UAVs must be able to communicate wirelessly between them. Hence, in this work, we propose a strategy which allows UAVs to efficiently communicate with each other. In particular, we propose a swarm management protocol that integrates mission control, message relaying and topology discovery to achieve its goal with maximum effectiveness. In addition, a minimal message overhead should be introduced by the protocol considering the limitations of the wireless channel, and also to provide more bandwidth to the actual applications running over the UAV swarm. Performance results obtained in a realistic UAV emulation environment show that the proposed approach is able to find the topology graph in just a few seconds, and that the overall mission time associated to missions involving multi-hop relaying (greater distances between UAVs) experiences a very moderate increase compared to a standard mission.