Speaker
Description
Non-cooperative rendezvous and proximity operations (RPO) are becoming a central challenge in orbital mechanics due to the rapid growth of space traffic and debris, as well as the increasing demand for on-orbit servicing, inspection, and active debris removal missions. Unlike cooperative scenarios, non-cooperative targets do not provide navigation data, attitude information, or dedicated docking interfaces, which significantly increases the complexity of guidance, navigation, and control.
This work investigates a geometric initialization strategy for indirect optimal control applied to orbital transfers. The trajectory design problem is formulated within the framework of the Pontryagin Maximum Principle, which leads to a two-point boundary value problem involving both state and adjoint variables. While indirect methods provide highly accurate and fuel-efficient solutions, they are well known to be extremely sensitive to the initialization of the costates.
To address this difficulty, a Lyapunov-based feedback law is first constructed to generate a dynamically feasible trajectory that converges toward a target orbit. This stabilizing trajectory is then used as an initial guess for the indirect optimal control solver. In particular, the Lyapunov structure provides a geometrically meaningful approximation of the costate direction, enabling a more robust initialization of the shooting method.
The approach is currently investigated on a simplified problem corresponding to a launcher upper-stage insertion toward Geostationary Transfer Orbit. This controlled test case allows us to analyze the relationship between Lyapunov-based stabilization and optimal costate trajectories before extending the framework to more complex scenarios, including three-dimensional dynamics and non-cooperative RPO missions.