Designing time-optimal trajectories for solar sail spacecraft is highly difficult due to the strong nonlinearity of the solar radiation pressure model, the attitude–orbit coupling, and the prevalence of local minima, which often lead to the failure of gradient-based solvers lacking adequate initial guesses. To address this issue, the present study proposes a two-stage hybrid optimization framework: a coarse-search stage uses B-spline-parameterized metaheuristics to identify a dynamically feasible trajectory, which then serves as a physics-informed warm start for direct-collocation-based gradient refinement, strictly satisfying the full nonlinear dynamics and terminal constraints. The methodology is validated through three increasingly difficult space missions: a rendezvous between Earth and Mars (Case A), a near-Earth asteroid rendezvous considering non-ideal optical reflection (Case B), and a Solar Polar Orbiter mission necessitating an 82.75 inclination adjustment with a thermal safety constraint (Case C). Statistical evaluations reveal that no single metaheuristic dominates universally; each algorithm’s suitability is contingent on the problem’s constraint structure. The hybrid framework further shows a consistent advantage over stand-alone direct collocation: by redirecting the gradient solver toward favorable convergence basins, it locates transfer solutions that remain inaccessible from a cold start.
He et al. (2026) studied this question.