Abstract: Aiming at the problem that multiple performance indicators need to be considered comprehensively for the trajectory planning of the future lunar surface landing power descent，this paper proposes a multi objective trajectory planning method that optimizes first and decides later. First, the multiple metrics of trajectory planning are decomposed under the framework of MOEA/D-AWA (multi objective evolutionary algorithm based on decomposition with adaptive weight adjustment) to obtain several single indicator sub-problems. After that, the convex optimization algorithm is designed as the underlying algorithm for solving the single objective sub-problems. After iterative optimization, a set of powered descent segment flight trajectories can be obtained, which constitutes the Pareto optimal solution set. Finally, according to the fuzzy decision theory, the multiple trajectory indicators corresponding to each Pareto optimal solution are gradually downgraded and evaluated comprehensively, and the flight trajectory under multi-indicator constraint is obtained after the decision. The simulation results show that the trajectory planning method can optimize a set of power descent trajectories under the integrated multi objective situation, and can decide the optimal power descent trajectories from them according to different mission requirements, which can effectively solve the multi objective trajectory planning problem of the lunar vehicle.

Key words: lunar vehicle, trajectory optimization, multi objective optimization, Pareto frontier, fuzzy decision