Deakin University
Browse

File(s) not publicly available

MACHINE LEARNING FOR PREDICTING THE OUTCOME OF TERMINAL BALLISTICS EVENTS

conference contribution
posted on 2024-03-14, 05:13 authored by Shannon RyanShannon Ryan, NM Sushma, AAV Kumar, J Berk, T Hashem, Santu RanaSantu Rana, Svetha VenkateshSvetha Venkatesh
Machine learning (ML) is well suited for the prediction of high-complexity, high-dimensional problems such as those encountered in terminal ballistics. We evaluate the performance of ML-based regression models on two common terminal ballistics’ problems: (a) predicting the V50 ballistic limit of monolithic metallic armour impacted by small and medium calibre projectiles and fragments, and (b) predicting the depth to which a projectile will penetrate a target of semi-infinite thickness. The models provide excellent agreement with test data. Although extrapolation is not advisable for ML-based regression models, for applications such as lethality/survivability analysis, such capability is required. To enable extrapolation, we implement expert knowledge and physics-based models via enforced monotonicity, as a Gaussian prior mean, and through a modified loss function. The physics-informed models demonstrate improved performance over both classical physics-based models and the basic ML regression models, providing an ability to accurately fit experimental data when it is available and then revert to the physics-based model when not.

History

Volume

2

Pagination

2043-2045

Location

Bruges, Belgium

Start date

2023-10-16

End date

2023-10-20

ISBN-13

9781605956923

Language

eng

Title of proceedings

Proceedings - 33rd International Symposium on Ballistics, BALLISTICS 2023

Event

Ballistics. Symposium (2023 : 33rd : Bruges, Belgium)

Publisher

DEStech Publications

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC