Target maneuver trajectory prediction based on RBF neural network optimized by hybrid algorithm

Target maneuver trajectory prediction plays an important role in air combat situation awareness and threat assessment. To solve the problem of low accuracy traditional method model, a target model based on phase space reconstruction-radial basis function (PSR-RBF) neural network is established by combining characteristics with time continuity. In order to further improve performance rival penalized competitive learning (RPCL) algorithm introduced determine structure RBF, Levenberg-Marquardt (LM) hybrid improved particle swarm optimization (IPSO) k-means are optimize parameter PSR-RBF constructed. An independent 3D coordinates proposed, manuver sample data constructed using training selected instrument (ACMI), established. verify precision real-time simulation experiment performed. The results show that better, proposed better. confirms effectiveness applicability model.