Previous Abstract Return to Session F11 Next Abstract

Session F11: Modeling and Simulation: Results of Strategic Studies

The Development and Demonstration of the Weapon Effectiveness Database
Ezra Bregin and Lauren MacIver, JHU/APL
Location: Ballroom D
Date/Time: Wednesday, Jun. 4, 4:45 p.m.

Traditional methods for assessing the effectiveness of a precision-guided munition (PGM) typically rely on simulation tools that consider a limited set of trajectories and target types. To overcome these limitations, JHU/APL has developed a scalable simulation tool built on SWAC’s High-Fidelity PNT simulation framework. This tool computes the probability of hitting (Ph) over an area of interest within a specified threat environment by leveraging the independence of multiple trajectories and high-performance computing resources.
In this work, we detail the development of a tool designed to generate a geometrically distributed weapon effectiveness database (WED) using SWAC’s High-Fidelity PNT framework. This framework integrates components such as Controlled Reception Pattern Antennas (CRPA), Inertial Navigation Systems (INS), and a broad array of GNSS signals. We present an overview of the PNT simulation framework and demonstrate its application by constructing a comprehensive WED. Each database entry is structured as a two-dimensional table (latitude and longitude corresponding to a target location) populated with corresponding Ph values and is defined for specific launch locations, weapon models, and threat configurations. A given weapons model may include CONOPS, a CRPA (ranging from two to fourteen elements), MEMS-to-navigation grade IMUs, a 12-state INS, seeker definition, 6DOF rigid body dynamics, and guidance algorithms. We also present methods for non-uniform sampling that allow for generating a larger, more valid database without increased computing resources.
The WED supports multiple use cases, including:
1. Evaluating the Impact of PNT Solutions: Assessing alternatives such as the rapid deployment of Regional Military Protection (RMP)
2. Enhancing Mission Planning: Informing commanders’ decisions on the optimal deployment of launch platforms.
3. Enabling Rapid Digital Wargaming: Providing high-fidelity model approximations where real-time full simulations would be computationally prohibitive.
4. Accelerating Reinforcement Learning for Campaign Management: Mitigating the parameter explosion and slow convergence issues inherent in using direct high-fidelity simulation inputs through precomputed tables.
We illustrate these applications with detailed examples and discuss the advantages of integrating precomputed high-fidelity data into operational and decision-support systems.

Previous Abstract Return to Session F11 Next Abstract