Ground thread identification of the reconnaissance and strike integrated UAV based on improved Direct Inference algorithm

Ground thread identification is the foundation of tactical decision-making to the reconnaissance and strike integrated UAV (Unmanned Aerial Vehicle). Based on the network structure of a single time slice, the traditional Direct-Inference (DI) algorithm is improved for fuzzy discrete dynamic Bayesian network model in this paper. First of all, the improved DI algorithm is derived, and then the network model based on this algorithm for the purpose of UAV ground thread identification is established as an example. The simulation results demonstrate that the algorithm can not only accurately infer the type of the ground thread by combining feature information, but also significantly shorten the cost time of the Inference.