A Stochastic Grammar Approach to Predict Flight Phases of a Hypersonic Glide Vehicle

Hypersonic glide vehicles (HGVs) fly at very high velocities and demonstrate high agility, which creates challenges in forecasting their flight behavior (e.g., aerodynamics, flight paths, maneuvers). In this paper, we describe a method for predicting future flight phases of a HGV. Flight phases could be modeled as categorical labels that correspond to different types of flight behavior. For our purposes, we define flight phases based on magnitude of rate of change of total energy (i. e., summation of kinetic and potential energy) of a vehicle. We use methods from natural language processing to model the flight phases as “words” and the HGV trajectories as “sentences.” We learn a “grammar” from the HGV trajectories, which we use for our prediction task. Given “words” from the initial part of a HGV trajectory and the “grammar”, we can predict future “words” in the “sentence” (i.e., future HGV flight phases in the trajectory). We demonstrate that this approach successfully predicts future flight phases for HGV trajectories, especially in scenarios with limited training data.