College of Automation Engineering，Nanjing University of Aeronautics and Astronautics
Artificial intelligence + special project of basic scientific research in Central University（NZ2020003）
Based on the flight characteristics and the control strategy of UAV with fixed wings, temporal network and cellular automata theory are used to analyze the failure influence mechanism of swarm UAV in the case of malicious attack to controller. Firstly, the change of the topology network of swarm UAV is studied by temporal network, and a method based on hop number is then proposed to determine the fault propagation path. Secondly, utilizing the status information of swarm UAV, the cellular object that satisfy the characteristics of swarm UAV is established. Based on the principle of local information interaction, the state conversion rules is determined so that the dynamic change of the influence degree of UAV failure can be identified, by which the weights of failure influence are determined by the vector projection method according to the influence of the neighbour information on the control law of a UAV. Finally, after building a simulation model, the prediction results compared with the actual fault influence degree are obtained via DCG algorithm and model distance, which verifies the effectiveness of the proposed fault influence model.