For reasons such as researching the origin of the universe, seizing space resources, and seeking new habitats, human exploration has gradually shifted towards deep space. Because of the proximity to Earth and unique gravitational characteristics, the Moon naturally becomes the first destination for deep space exploration. With the second wave of lunar exploration sweeping the globe, the lunar South Pole has become the primary destination due to its unique geographical terrain and strong resource potential. This work proposes a lunar constellation design method based on multi-objective optimization. Based on lunar frozen orbits and considering the presence of the Queqiao-2 satellite, four performance metrics representing coverage and navigation are selected as optimization objectives. Several representative constellation schemes are extracted from the Pareto front, and their configurations and performances are analyzed.