Visually Detecting Drones in Drone Swarm Formations Topologies

In drone swarm network, the topology is an important characteristic of the swarm and the level of bonding with each node is strongly influenced by the communication links between drones. Understanding the network structure of a swarm by recognising, monitoring, and intervening with nodes is important to understand the behaviour and intention of the swarms. Current methods for understanding drone organisation generally require access to the swarm network which is not practical, therefore, passive approaches for network topology estimation are needed. The research presented in this paper proposes an approach using computer vision to detect drones in distinct swarm formations. The approach is evaluated by simulation using multiple different drone formations. It is anticipated that the results of this study will lead to approaches for detecting and understanding different drone formations.