Distributed Consensus-Based Loitering Control for Open Multi-Agent Drone Swarms

This paper presents a fully distributed framework for loitering control in Open Multi-Agent Systems (OMAS) modeling a drone swarm, where agents can dynamically join or leave the network. This approach, which operates without requiring any global knowledge or centralized coordination, consists of two tightly integrated components: i) a distributed protocol enabling each agent to estimate the time-varying swarm centroid using only local interactions; and ii) a vector-field-based control law that leverages this local estimate to generate stable orbital motion while ensuring collision avoidance via decentralized repulsive terms.