Hypersonic Vehicle Trajectory Classification Using Convolutional Neural Network

Abstract

This paper proposes a Hypersonic Convolutional Neural Network (HCNN) for hypersonic aircraft classification based on a vehicle flight path. The experiments on synthetic data show that HCNN has a high-resolution of identification capability on three types of targets (ballistic missile, hypersonic glide vehicle, hypersonic cruise missile) even in conditions of increased measuring errors, increased time step, and with no preliminary primary data processing. Moreover, one can easily adjust the proposed network for discrimination of an enlarged number of vehicle types.