The　low-energy　Moon　return　trajectory　is　of　great　importance　for　lunar　exploration,　especially　for　the　sample　retun　mission.　In　this　paper,　a　new　GPM/VSM　strategy　is　proposed　to　finish　the　return　trajectory　with　finite　thrust.　This　new　strategy　not　only　can　simplify　the　optimization　by　converting　the　trajectory　maneuver　into　a　rendezvous　problem　but　also　guarantee　the　convergence　of　the　algorithm.　Compared　with　traditional　methods,　the　new　GPM/VSM　strategy　only　evolves　the　integration　of　the　rendezvous　part,　which　reduces　the　amount　of　calculation　significantly.　What＇s　more,　using　the　GPM　algorithm　to　instead　of　the　maximum　principle　algorithm　avoids　the　guess　for　co-state.　The　simulation　results　show　that　the　GPM/VSM　can　offer　results　with　high　accuracy　and　efficiency　and　the　final　error　is　small　enough　to　guarantee　the　successful　return.