Generalized Framework for Cislunar Satellite Placement and Operations Optimization

Expanding infrastructure and traffic in the region between the Earth and Moon necessitates commensurate commitments in monitoring and surveillance architectures. This work presents a formulation for optimizing the placement of space assets via a heirarchal optimization scheme. The outer stage leverages first order methods to optimize asset placement in a continuous space, while the inner stage optimizes operation-level decisions; the gradient information generated from the inner loop is used in the outer loop for iterations. We show that this formulation, unlike previous approaches, is modular, allowing researchers to optimize a constellation's phasing while remaining agnostic to the underlying optimzation signal. We show its applicability on two application problems, namely the sensor tasking and communication packet routing problems. This formulation can be easily generalized to any dynamical system with periodic orbits in state space, which yields potential application outside of the space context.