Seeking the Scalability in Algebraic Graph Based Unmanned Aerial Vehicle Formation Control

This paper studies the scalability problem for multi-UAV formation system governed by double-integrator dynamics. More specifically, we focus how to build communication links with fixed control parameters such that the formation can always keep stable when adding arbitrary number of UAVs. A bio-inspired method - Veteran Rule is proposed to solve this problem. Compared to the existing methods, our proposed method does not require to re-design or adaptively adjust the control parameters/gains for the changed Laplacian matrix due to adding new UAVs in the formation system. Furthermore, the optimal convergence rate can be easily designed in our method. Surprisingly, we find that the optimal convergence rate is reached when all the in-degrees equal a particular value, rather than goes unboundedly. Finally, simulation results corroborate the effectiveness of our results.