In the three-body game scenario,it is necessary to predict the trajectory of incoming targets to enhance the interception capabilities of defensive missiles.Limited by the transmission rate of data link,the update frequency of target information is low,and the trajectory prediction method based on the Kalman filter and trajectory fitting is not applicable.Therefore,this study presents a target trajectory prediction method capable of classification under low information support.Firstly,the trajectory database of incoming targets was established based on the detection ability of airborne radar,and then the classification neural network and trajectory prediction neural network were trained using the trajectory database.The type of the target was determined online according to the target information transmitted by the data link,and after that,the optimal estimation of the initial state of the incoming target was acquired by the least square method.Finally,the trajectory prediction was realized.Simulation results show that the proposed method can successfully achieve high-precision trajectory prediction under low information support condition.
low information support conditiontrajectory predictionneural networkleast square methodthree-body game
维普
万方数据
