Physics-informed Conditional Normalizing Flows for Angles-only Cislunar Orbit Determination
For astrodynamics practitioners, this offers a data-driven approach to improve orbit determination in the challenging cislunar regime, though the method is incremental over existing generative modeling techniques.
This work applies conditional normalizing flows to estimate the probability distribution of initial spacecraft states from angles-only cislunar observations, enabling multimodal posterior sampling that provides competitive warm starts for classical orbit determination algorithms.
Generative Astrodynamics is advanced in this work by extending generative modelling to an orbit determination problem in the cislunar environment. The task is formulated as conditional density estimation, aiming to infer the probability distribution of the initial state from angles-only measurements over short observation arcs. A normalising flow is trained on perturbed topocentric observations from Near Rectilinear Halo Orbits, enabling a flexible and potentially multimodal posterior representation. Given new measurements, the learned density is sampled to generate statistically consistent and physics-informed state hypotheses. These estimates are refined via nonlinear least-squares minimisation, providing a competitive warm start for classical algorithms.