Unified Trajectory Optimization Method for Multipass Aeroassisted Maneuvers via Convex Optimization

Aeroassisted maneuvers, including aerocapture, aerobraking, and aeroassisted orbital transfers, represent a novel method for orbital maneuvers due to their significant low-fuel advantages in reducing orbital energy. This paper proposes a unified trajectory optimization method based on sequential convex optimization for multipass aeroassisted maneuvers. We initially reconstruct a new independent variable to reduce the dimensionality of dynamics. This new variable not only overcomes the potential non-monotonicity issues associated with traditional energy-based independent variables but also significantly enhances the convergence of sequential iterations. For the nonconvex nonlinear equality constraints posed by interior point connection conditions and endpoint maneuvers, we propose a relaxation strategy based on an adaptive rectangular trust region, ensuring feasibility during iterations. Additionally, to enhance the relaxation factors for convergence and ensure the activation of bank angle constraints under multiple constraints, we introduce a dynamic adaptive penalty strategy, ensuring the algorithm’s adaptability across multiple scenarios. Furthermore, we establish feasibility criteria and initial trajectory generation strategies to maximize the autonomy of the trajectory optimization. Simulation analyses using lunar return and Mars exploration missions demonstrate the effectivity of the proposed method and validate its strong robustness across varying initial orbits (spanning hyperbolic orbit to medium orbit).