Abstract: In order to solve the problem of collision of Unmanned Aerial Vehicles (UAV) in a local airspace and the possibility of chain collision, innovatively based on the theory of complex networks, the key node selection and the sense selection are applied, maximizing the security of the threat to the UAV group. By analyzing the status information of the UAV group, the most important UAV (key nodes) is selected to avoid collisions, and at the same time, the robustness minimum principle is adopted to select the collision avoidance direction. Simulation results of the two typical UAV flight cases show that this strategy can not only effectively solve the current conflict problem of UAVs, but also prevent chain collisions and achieve overall optimization. Quantitative simulation experiments are conducted to validate the feasibility and scalability of the proposed algorithm. Compared with the random choose direction collision algorithm, the results show that this algorithm can indeed improve the safety of the UAV group.

Key words: multi-UAV, conflict resolution, complex network, key nodes, robustness