Multi-Objective Constrained Constellation Design and Characterization for Cislunar Space Object Tracking

Space-based observer constellations can circumvent the shortfalls of ground-based observations. With a greater proliferation of space objects in cislunar space anticipated in the near future, custody retention, object registration, and tracking become paramount for safe and secure operations. Determination of tracking feasibility requires an in-depth analysis of sensor models, target observability, onboard resources, and observation geometry in tandem. While a globally optimal configuration can be sought, the solution space is littered with local minima, and the design space is infinite. This work introduces an approach to optimize the initial configuration of an observer constellation to track a cislunar resident space object target exhibiting two distinct behaviors: 1) coasting on a periodic orbit in the Earth–Moon CR3BP and 2) performing a low-thrust transfer between two periodic orbits in the Earth–Moon CR3BP. Angles-only and angles-with-range measurement models are considered, and a genetic algorithm is used to optimize constellations in a systematically filtered design space. Multi-objective optimization results are also presented to map the Pareto tradeoffs between minimizing estimation error and uncertainty. Monte Carlo analysis is conducted to explore the local space around the GA solution and explore optimality. The approach results in interesting multi-spacecraft constellation architectures while also characterizing the Pareto tradeoffs. Further contextualization of the presented architectures is carried out by evaluating robustness to uncertain initial conditions, sensitivity to initial phase, and extended evaluation times for both scenarios.