arXiv:2606.30936v1 Announce Type: new Abstract: 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
The increasing focus on cislunar space for scientific, commercial, and national security interests necessitates advanced, AI-driven solutions for precise orbit determination, especially with limited sensor data.
This development enhances space domain awareness and operational capabilities in a critical and increasingly congested cislunar environment, impacting future space infrastructure and potential resource utilization.
The ability to accurately track objects with angles-only measurements and short observation arcs improves resilience and autonomy in space operations, reducing reliance on conventional, more data-intensive methods.
- · Space agencies
- · Satellite operators
- · AI/ML developers
- · Cislunar economy stakeholders
- · Traditional orbit determination methods
- · Countries with limited space surveillance capabilities
Improved precision in tracking objects in the cislunar space reduces collision risks and optimizes mission planning.
Enhanced cislunar capabilities could accelerate development of lunar infrastructure and resource extraction, increasing geopolitical competition for celestial bodies.
The application of generative AI to astrodynamics could lead to fully autonomous space traffic management systems, fundamentally altering how space operations are conducted.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG