Scheduling Tasks for AFSCN Resources using MBO Autonomous Planning (STARMAP)

ABSTRACT:Scheduling Air Force Satellite Control Network (AFSCN) resources is challenging because windows of visibility are limited between satellites and remote tracking stations on the ground, and there are more requests for access than available resources. Human scheduling at a centralized location is difficult, time-consuming, and inefficient because it is filled with complex unwritten rules. AFSCN schedulers need an intelligent, distributed, real-time planning and scheduling tool that uses spacecraft telemetry data to automatically allocate antennas based on complex optimization criteria to reduce the burden on human schedulers and enable them to focus on higher level objectives. Charles River Analytics proposes to develop a decision support tool for Scheduling Tasks for AFSCN Resources using MBO Autonomous Planning (STARMAP). STARMAP is an intelligent, real-time planning and scheduling framework for distributed market-based optimization (MBO) of antenna resources to improve the scheduling effectiveness of satellite networks. Our approach involves modeling satellites, tracking station facilities, and customer agencies as dynamic agents that, along with custom arbitrage agents, optimize the global schedule in a continuous, real-time manner that is robust to external complications and can effectively incorporate both high and low priority contact request tasks into a globally optimal schedule.BENEFIT:STARMAP will provide immediate benefits to the AFSCN enterprise, the 22 SOPS, and the SOCs by enabling intelligent, distributed, contact request scheduling that improves the effectiveness of satellite networks, and enables decreased staffing by AFSCN schedulers. We will enhance our Dynamic Rendering and Information Visualization Engine (DRIVE) with basic animation, agent integration, and orbital visualization capabilities.