Bayesian optimization for the spanwise oscillation of a gliding flat plate

Bayesian optimization is a powerful tool in the kinematic design of bio-inspired flight, where expensive high-fidelity numerical simulations are required. However, the traditional single acquisition strategy cannot adapt to all design problems, which have various input-output characteristics. The hybrid acquisition strategy improves the robustness by utilizing three types of acquisition functions, and each acquisition function is assigned three different balance parameters. The kinematics of a gliding flat plate that oscillates along the spanwise direction has been optimized to enhance the power efficiency by using the Bayesian optimization method. The power factor under the optimal spanwise oscillation is 1.97 times as large as that without spanwise oscillation, which indicates the good capability of the optimization method for bio-inspired locomotion.