Numerical exploration of low lunar, long-lifetime orbits

Low lunar orbits that persist over a long duration without impacting the Moon continue to interest the astrodynamics community for trajectory design and space situational awareness. To support exploratory analysis of these low lunar, long-lifetime trajectories in a high-fidelity dynamical model, this paper uses a clustering-based approach. First, low lunar trajectories are numerically generated from coarsely sampled initial conditions for up to 180 days in a 100 $\times$ 100 lunar gravity model along with the point mass gravity of the Earth and Sun. Each trajectory is represented by a subset of its perilunes in a Moon-fixed frame. Distributed trajectory clustering is then used to group trajectories with a similar evolution of eccentricity and argument of perilune, regardless of their initial phasing. From selected global clusters, trajectories with a low level of drift in their evolution of perilune over 180 days and geometrically distinct perilune groundtracks are isolated as long-lifetime orbits. Sets of geometrically similar long-lifetime orbits are presented, analyzed, and then coarsely compared to analytically derived frozen orbit predictions from the existing literature.

The online version contains supplementary material available at https://doi.org/10.1007/s10569-026-10334-x.