Optimizing Surveillance Satellites for the Synthetic Theater Operations Research Model

TL;DR

This paper presents a mixed-integer linear optimization model for directing surveillance satellites in theater-level conflict scenarios, significantly improving search coverage over existing heuristics.

Contribution

It introduces a novel optimization framework that prescribes satellite and sensor operations, handling large-scale instances with millions of variables and constraints.

Findings

55% improvement in search coverage over heuristics

Successfully solves large-scale instances with 22 million variables

Enhances the fidelity and efficiency of satellite search planning

Abstract

The Synthetic Theater Operations Research Model (STORM) simulates theater-level conflict and requires inputs about utilization of surveillance satellites to search large geographical areas. We develop a mixed-integer linear optimization model that prescribes plans for how satellites and their sensors should be directed to best search an area of operations. It also specifies the resolution levels employed by the sensors to ensure a suitable fidelity of the resulting images. We solve large-scale instances of the model involving up to 22 million variables and 11 million constraints in scenarios derived from STORM. On average, the model yields 55% improvement in search coverage relative to an existing heuristic algorithm in STORM.