Adaptive decentralized flocking control of multi-UAV circular formations based on vector fields and backstepping.

This is a paper on controlling fixed-wing unmanned aerial vehicle (UAV) swarm formations while coordinating their flocking to a specified circular path. The proposed non-uniform in both magnitude and direction path-following vector fields enable the aircraft of the entire group to converge to a circular motion around a target while also attaining and maintaining relative phase-shift angles between the UAVs. It is thereby assumed that UAVs use decentralized consensus for their neighbor-neighbor​ coordination, which implies unconstrained scalability of the formation. The highlight of this research is that it gets rid of the conventional assumption that all the UAVs must initially be on a circular path and follow it strictly, which makes the proposed approach more practical. The obtained backstepping-based control commands explicitly factor in the input constraints and make the UAV course angles and speeds converge to the vector field-specified values. The inevitable parameter uncertainties of UAV kinematic models can destabilize the formation, which is why adaptive self-tuning is applied. The new decentralized UAV flocking controller has been tested by detailed numerical MATLAB/Simulink experiments, including comparative experimentation, using realistic six degree-of-freedom (DoF) 12-state nonlinear UAV models; numerical modeling demonstrates the proposed approach stable for a variety of initial conditions.