Sensor Task Optimization and Real-time Management (STORM)

Sensor management and, more specifically, the capability to automate scheduling and dynamically re-task sensor assets in light of changing operational requirements and mission objectives is a crucial factor for ensuring information dominance for ourwarfighters. Here, we propose a Sensor Task Optimization and Real-time Management (STORM) system for the scheduling, coordination, and path planning of heterogeneous sensor assets and the dynamic adaptation and replanning of those assets in response tochanging battlespace conditions. The system uses GIS processing along with target and threat prediction to prioritize the area to be searched. Scheduling is accomplished using Ant Colony Optimization (ACO), which is an agent-based approach thatincorporates heterogeneous sensors and a high-fidelity model of the mission environment. Plan execution monitoring is used in conjunction with ACO to make real-time adjustments to the plan to account for changing and unexpected conditions.For Phase II we will: 1) enhance the Phase I system; 2) expand functionality to address replanning; 3) implement a full-scope prototype; 4) demonstrate performance across multiple tactical scenarios; 5) provide a transition plan for follow-on systemintegration as a GCCS-M segment for sensor management decision support; and 6) and determine requirements for commercialization. Commercial applications of the proposed approach to schedule optimization exist for a wide variety of domains including transportation, airways and railway time-tabling, and inventory control. In addition, the proposed effort will impact the developmentand enhancement of our Intelligent Agent Toolkit (IAT) product, via the incorporation of the adaptation and scheduling components of the proposed STORM developed under this SBIR effort.