Optimization and valuation of recongurable satellite constellations under uncertainty

Space-based persistent surveillance provides decision makers with information necessary to effectively respond to both natural and man-made crises. This thesis investigates a reconfigurable constellation strategy that utilizes on-demand, maneuverable satellites to provide focused regional coverage with short revisit times at greatly decreased cost when compared to traditional static satellite constellations. The thesis develops and demonstrates a general framework to guide the design and optimization of reconfigurable satellite constellations specifically tailored to stakeholder objectives while considering requirement uncertainty. The framework is novel in that it avoids many of the assumptions and simplifications of past research by: 1. explicitly considering uncertainty in future operating conditions; 2. concurrently optimizing constellation pattern design, satellite design, and operations design; and, 3. investigating layered and asymmetric patterns. The framework consists of three elements: a detailed simulation model to compute constellation performance and cost for a variety of architectures and patterns, Monte Carlo simulation to determine how well each design performs under uncertain future conditions, and a parallel multi-objective evolutionary algorithm developed from the -NSGA-II genetic algorithm to find designs that maximize performance while simultaneously minimizing cost. Additionally, a new performance metric is developed to measure directly how well a design meets desired temporal and spatial sampling requirements and a decision model and optimal assignment process is developed to determine how to employ the option of reconfigurability to respond to specific regional events. The framework was used to perform 85 optimization runs selected to compare the costeffectiveness of several constellation architectures over varied operating conditions and coverage requirements. All optimization runs were performed in less than three months, demonstrating that parallel computing coupled with sophisticated optimization routines enable rapid spiral development of satellite constellations. Results show that reconfigurable constellations cost 20 to 70% less than similarly performing static constellations for the scenarios studied. The cost savings grows with increasingly demanding coverage requirements. Results from optimizing a fully asymmetric constellation pattern led to two the development of new ‘quasi’-asymmetric patterns that were found to significantly outperform symmetric patterns