LSTM Network-Based Maneuvering Recognition and Long-Range Guidance Design for Intercepting Fighter Aircraft

A target maneuvering recognition algorithm based on long short-term memory (LSTM) network and a long-range guidance strategy law are proposed for intercepting highly maneuverable fighter aircraft. Firstly, the maneuvering modes of the highly maneuverable fighter aircraft are modeled, and an LSTM network is trained to recognize the fighter’s maneuver mode in real time from and kinematic features. Secondly, the recognized maneuver information is then used to predict a virtual interception point and to construct a midcourse guidance law, as well as to design a terminal guidance law. Finally, simulation results demonstrate that the proposed maneuver recognition method achieves high recognition accuracy, and that the long-range guidance law including midcourse law and terminal guidance law based on maneuvering recognition algorithm exhibits lower energy consumption and higher accuracy compared to the guidance law without maneuvering recognition information.