Sergey Aksenov

In the circular restricted three-body problem, families of periodic and quasi-periodic orbits exist in the vicinity of the libration points and are of significant interest for practical applications. This work presents a neural network-based algorithm for orbit construction and station-keeping, applicable to both periodic and quasi-periodic orbits near collinear libration points. The method employs a fully connected neural network trained on labeled trajectory data and iteratively refined using trajectories generated during previous stages of the algorithm. The proposed approach was applied to orbit generation near the Sun-Earth L1 libration point and demonstrated the ability to generalize to a broader class of trajectories than those included in the training set.

The circular restricted three-body problem is used as an approximate model in space mission planning. Its periodic solutions around equilibrium points, which are referred to as the libration points, are utilized for exploration of possible spacecraft trajectories in the preliminary stages of mission design. In this paper, a numerical methodology for periodic libration point orbits (LPOs) computation is introduced and applied to the construction and study of $N$-periodic (up to $N = 6$) quasi-planar orbit families in the Earth-Moon system. The stability and the bifurcation points of these families are determined. The proposed method is based on an iterative algorithm searching for the initial state vector of periodic LPOs, which allows computing unstable long-periodic and large-amplitude orbits. The method is suited to perform a straightforward switch to bifurcating branches of periodic orbits.