Fault Diagnosis of UAV Formations Communication Links Based on Belief Rule Base

This paper investigates the fault diagnosis of UAV communication links using Belief Rule Base (BRB). Firstly, a BRB-based fault diagnosis model is constructed. By leveraging expert knowledge, appropriate reference values and reference points are set for the premise attributes and output results, and belief rules are constructed and initial belief degrees are assigned to complete the construction of the BRB model architecture. Furthermore, considering the practical characteristics of UAV communication links, a communication topology graph and a communication weight matrix are constructed to model faults in detail, and the received signal is utilized as the input of the BRB model to diagnose the fault types. Then, to address the accuracy limitations caused by the subjectivity of expert knowledge, the P-CMA-ES algorithm is employed to optimize the initial BRB model. Finally, simulation experiments validate the model's effectiveness in diagnosing fault types and its applicability to practical scenarios.