<p>For reusable vehicles, traditional deterministic optimization techniques can no longer meet their design requirements. Against this background, the concept of uncertainty optimization has been proposed and has stimulated extensive research, among which the reliability-based robust design optimization (RBRDO) method is adopted in this paper due to its ability to improve the reliability and quality of the product at the same time. However, for complex engineering design problems, the RBRDO method faces the problems of low computational efficiency and difficulty in solving. In order to tackle these problems, this paper proposes a sequential sampling Kriging surrogate model RBRDO method, and takes the turbine blades in liquid rocket engine (LRE) as the research object to carry out the reliability-based robust design optimization for its anti-vibration fatigue performance. In order to verify the effectiveness of the proposed method, this paper firstly carries out resonant fatigue experiments on the initial turbine blade. Then, the probabilistic vibration fatigue life prediction is carried out according to the uncertainty factors in the actual experimental conditions of turbine blades. Finally, the vibration fatigue resistance of the turbine blade is optimized using the method proposed in this paper. The results show that the vibration fatigue life of the optimized turbine blade is improved by about 10% and the standard deviation is reduced by about 4%, which in turn proves the effectiveness of the method of this paper.</p>