Evolution of Formation Control Algorithms for Unmanned Aerial Vehicles

This paper explores existing approaches of Formation Control Algorithms for Unmanned Aerial Vehicles. Addressed approaches are mainly categorized as centralized and decentralized. Special attention is given to reviewing the methodology and evolution of Leader-Follower, Behavior-based, Virtual Structure, and Artificial Potential Field approaches to identify limitations and future works. Many of the existing work related to Formation Control Algorithms are developed considering constant environmental conditions and criteria. Hence it is envisaged that Formation Control Algorithms can be improved or new approaches can be proposed to achieve better performance in different weather conditions like fog, wind, thunderstorms, snow, and varying altitudes with dynamic obstacles. The capabilities of the approaches should be further explored by integrating deep learning and reinforcement learning in addition to Machine Learning and Artificial Intelligence.