Threat assessment for air-to-ground combat of UAVs using improved Bayesian networks

Abstract

Aiming at the problems of information uncertainty and real-time faced by ground threat assessment during UAV air-to-ground combat, the limitations of existing threat assessment methods are analyzed, and a new threat assessment method based on improved Bayesian network is proposed in this paper. In this method, a hybrid-driven approach of threat data and expert knowledge is used to improve the rationality of threat assessment model, and a layered threat assessment architecture is constructed. Threat levels are classified based on the relationship between carrier aircraft and threat envelope, threat radiation, etc. In order to improve the rationality of the ground threat assessment model, a calculation model of ground threat intervisibility probability is established, and the intervisibility probability is taken as the input of Bayesian network threat assessment model. Simulation results show that the model has less computation and good real-time performance, and the introduction of threat intervisibility probability improves the rationality of ground threat assessment.