Skeleton-Based Swarm Routing (SSR): Intelligent Smooth Routing for Dynamic UAV Networks

A swarm of unmanned aerial vehicles (UAVs) requires the transmission mission-related data across network. The resource constraints and dynamic nature bring critical challenges to design UAV routing protocols. Most conventional ad hoc schemes are not intelligent cannot adapt swarming networks. On other hand, some artificial intelligence (AI)-based may consume significant computational resources in UAVs. In this article, a low-cost, adaptive protocol, namely skeleton-based (SSR), is proposed, which exploits an online learning algorithm topology features mission-driven distribute traffic over optimal routes. Here, skeleton represents most stable parts formation. SSR architecture consists three modules: 1) geometric addressing module, assigns coordinates each node based on structure; 2) leaf-like pipe allows selection multiple candidate routes around shortest path; 3) An low-complexity model determines how packets inside achieve load-balanced, high-throughput transmissions. proposed scheme can also facilitate formation construction morphing. simulation results show that protocol noticeably improve network performance (up 100% throughput improvement) compared single path schemes, such as ad-hoc on-demand distance vector (AODV) link-quality traffic-load aware optimized link state (LTA-OLSR)