Operational Scheduling of Deep Space Radars for Resident Space Object Surveillance

Abstract

As space becomes increasingly congested and contested, new capabilities of rivals to threaten vital assets and exploit the area for military advantages make it more important than ever for the United States to proficiently track and monitor space traffic and debris. However, currently the system of radars used by the Department of Defense to track objects in deep space operates in a way that is labor intensive, uncoordinated, and inefficient. In this thesis, we address these issues by automating and coordinating the radar scheduling process. We consider several complex radar systems that operate in an asynchronous, distributed environment and target space objects with varying priority levels, time windows, arrival frequencies, and task mission requirements.

We develop a mixed integer program capable of intelligently distributing task requests and building radar slew plans in a way that aligns with user objectives and system characteristics. We solve the optimization problem repeatedly over time, all while receiving and incorporating updated information, new task requests, and available feedback throughout the planning process. We test our methodologies on various tactical military scenarios and show that an optimization-based approach allows us to maintain custody of more space objects, better prioritize high value objects, and reduce operating costs when compared to a baseline greedy algorithm. We conclude that an automatic, centralized way of scheduling is viable and beneficial for use in the Space Situational Awareness (SSA) mission.