← Back to list of papers of the 2024 EuroGNC conference

CEAS EuroGNC 2024

Trajectory Prediction for Missile Targets: A Probabilistic Approach Using Machine Learning
Marc Schneider Research Associate, University of Stuttgart, Institute of Flight Mechanics and Controls, 70569 Stuttgart, Germany.
Renato Loureiro Research Associate, University of Stuttgart, Institute of Flight Mechanics and Controls, 70569 Stuttgart, Germany.
Torbjørn Cunis Lecturer, University of Stuttgart, Institute of Flight Mechanics and Controls, 70569 Stuttgart, Germany.
Walter Fichter Professor, University of Stuttgart, Institute of Flight Mechanics and Controls, 70569 Stuttgart, Germany.
Abstract:
Knowledge about the future trajectory of the target is an essential part of the guidance and control of guided missiles. Most existing prediction models either implicitly or explicitly assume a deterministic behavior of the target. Using this prediction model, the guidance law computes the control commands for the missile. In reality, targets exhibit diverse maneuver possibilities, rendering prediction a stochastic problem. This paper introduces the application of Conditional Normalizing Flows, a data-driven probabilistic approach that leverages Machine Learning to predict the probability density function of the future position of the target. These transform a Gaussian distribution into the target's estimated position distribution based on time and additional parameters describing the target dynamics. Simulated scenarios illustrate the approach's utility for targets with stochastic and deterministic maneuvers, and for ballistic targets as a proof of concept for complex scenarios. Leveraging probabilistic prediction, our approach empowers guidance laws to compute suitable control commands for the missile. This research bridges the gap between deterministic predictions and the stochastic reality of target behavior, contributing to the advancement of missile guidance systems.
Keywords: Probabilistic Trajectory Prediction; Conditional Normalizing Flows; Generative Models; Guided Missiles; Stochastic Target Dynamics
 View PDF CEAS-GNC-2024-103


Marc Schneider, Renato Loureiro, Torbjørn Cunis, Walter Fichter: Trajectory Prediction for Missile Targets: A Probabilistic Approach Using Machine Learning. Proceedings of the 2024 CEAS EuroGNC conference. Bristol, UK. June 2024. CEAS-GNC-2024-103.
BibTeX entry:

@Incollection{CEAS-GNC-2024-103,
    author = {Schneider, Marc and Loureiro, Renato and Cunis, Torbj\o{}rn and Fichter, Walter},
    title = {Trajectory Prediction for Missile Targets: A Probabilistic Approach Using Machine Learning},
    booktitle = {Proceedings of the 2024 {CEAS EuroGNC} conference},
    address = {Bristol, UK},
    month = jun,
    year = {2024},
    note = {CEAS-GNC-2024-103}
}