An efficient global optimization (EGO) algorithm has been used in the geometry design and motion for a manta ray-like flapping wing to optimize the best swimming performance. To this end, we combine a three-dimensional simulation using STAR-CCM+ with this framework to generate the optimal flapping wing geometry that maximizes the lift-to-drag ratio. The drag and moment coefficients between the original and optimized flapping wings were further compared in a servo-driven towing tank to validate the optimization effects. In addition, by associating the force experiments in a water tunnel, we successfully predict the motion trajectory for the largest thrust with the same EGO algorithm. A free-swimming test is carried out to verify the optimization. This study gives a unified scheme design optimization for the next-generation manta ray robot.