Contents I. Introduction
Interest in Unmanned Aerial Vehicles (UAVs) and swarms of UAVs has constantly been increasing in the past decade. The collective dynamics of swarm systems are not only a fascinating natural phenomenon by themselves but also have various engineering applications. Potential applications include search and rescue operations, surveillance/reconnaissance, exploration, and defence, to name a few. Tracking and capturing/enclosing a target is also a significant problem that may arise in some of the mentioned applications [2]. The objective of target capturing by a swarm of agents can be stated as to achieve a formation enclosing or surrounding a specific area or an object (a possibly moving target) by using local information [3]. One difficulty with the capturing problem could be that the agents may have access to only imperfect information about the target. In [1], a target capturing strategy for swarms composed of double integrator agents was developed assuming uncertainties on the target velocity and acceleration. This paper further extends and implements this work on a swarm of quadrotor UAVs. We perform simulations in MATLAB and Robot Operating System (ROS) environments, utilizing generic UAV models and realistic assumptions.
