Leader-following flocking for unmanned aerial vehicle swarm with distributed topology control

Abstract

To address the flocking issues of an unmanned aerial vehicle (UAV) swarm operating at a leader-follower mode, distributed control protocols comprising both kinetic controller and topology control algorithm must be implemented. For flocking the UAV swarm, a distributed control-input method is required forboth maintaining a relatively steady state between neighboring vehicles (including velocity matching and distance maintenance) and avoiding vehicle-to-vehicle collision. Furthermore, the stability of control protocols should be analyzed using the potential energy function. In particular, a distributed $\beta~$-angle test (BAT) rule in the proposed topology-control issue may allow each UAV to determine its neighboring set by exploiting the locally sensed information, thereby significantly reducing the communication overhead of the entire swarm. In addition, node-degree bound is derived to demonstrate the feasibility of the proposed algorithm, in which the optimal value in terms of convergence is analyzed. The flocking of the flying ad-hoc network (FANET) can be achieved in a self-organizing way without the use of an external control center via the distributed control protocols. Ultimately, the proposed analysis is verified by numerical results.