Distributed Spatiotemporal Optimization for Radar Mission Coordination using Cooperative Control

PI: Derek A. Paley

In support of the Missile Defense FNC Program, Force Level Radar Resource Management for IAMD project, the Collective Dynamics and Control Laboratory at the University of Maryland has contributed to the design of the conceptual architecture for dynamic management of radar tasking among multiple forces by applying tools from optimization and control to address the problem of dynamic, autonomous sensor coordination to maximize the collective sensor search volume and concurrently optimize the sensor-to-target assignment for long-range and short-range targets. This report describes the first year of a multi-year effort. Project deliverables include:

• A mathematical modeling framework for ship placement and radar resource allocation
• Analytic and numerical techniques to identify optimal strategies within the parameter space
• Metrics and candidate scenarios for algorithm evaluation

Background
The Office of Naval Research is conducting several science and technology projects as a part of the Sea Shield FNC MD Program, which is focused on improving active missile defense.  In a high threat-raid density environment, management of sensor resources supporting surveillance, tracking, combat identification, engagement, and kill assessment is critical to maximize the likelihood of defeating all threats targeting defended assets.  Current known capabilities are limited to setting sensor doctrine and manual engagement support coordination. Technical challenges to sensor resource management across the force include: variations in sensor capabilities; variations in sensor interfaces and controllability; how to define force level sensor support tasks; how to prioritize sensor support tasks; how to coordinate sensor support tasks; variations in sensor location (surface or airborne); differences between ballistic and airborne threat trajectories and threat complex diversity, and multiple sensor bands, accuracies, and update rates (e.g., rotating versus phased-array).  Furthermore, this coordination may be constrained by available communications links, but must degrade gracefully relative to a nominal communications environment.

Accomplishments
The University of Maryland Collective Dynamics and Controls Laboratory’s primary responsibilities include constructing a mathematical modeling framework for ship placement and radar resource allocation and utilizing analytic and numerical techniques to identify optimal strategies with the model parameter space. Future work will entail development of a distributed spatiotemporal optimization algorithm.  The contributions of this year were (1) we provided a nonlinear, two-dimensional model of multiple multifunction radar systems that concurrently search for and track targets within a specified area, subject to a radar resource constraint; (2) we solved the problem of maximizing the collective search area of multiple multifunction radar sensor systems while concurrently tracking a prescribed number of targets within that area; and (3) we posed and solved a novel target-to-radar assignment problem for coordinated radars that balances the radar tasking.  While the item's focus is on radar sensors, we believe its contributions may be of interest to a broad community seeking an approach to optimal sensor placement and target tracking for linear and nonlinear sensors in a resource-constrained environment. All of the algorithms were implemented in real time assuming minimal network delays and can be run on a standard laptop computer.