Abstract: Aiming at the multi-objective, nonlinearity and fuzziness of optimization of the rotary transplanting mechanism with planetary gear system, multi-objective optimization model of kinematics parameter was established based on satisfactory theory. Fuzzy comprehensive evaluation was used to quantify the kinematics performance of the transplanting mechanism. BP neural network was trained to build the mapping relationship of satisfactory degree and satisfactory function. Optimal solution and its evaluation were obtained by real-code genetic elitism strategy algorithms as follows: semi-major axis of the elliptic gear a was 18.10 mm; the ratio of semi-minor axis to semi-major axis of the elliptical gear k was 0.988; initial settling angle of the planting arm α0 was -42.56°; initial settling angle of the planet gear δ0 was11.56°; initial settling angle of the planet carrier φ0was 31.02°; the distance between the planet gear and the seedling needle tip S was 153.79 mm; and satisfactory degree was 93.11. The results show that the method can improve the efficiency and quality of design and meet the demand of designers and users further.