Performance enhancement of a bionic rigid-flexible coupling flapping wing based on composite learning control

The bionic rigid-flexible coupling flapping wing (BRFCFW) is inspired by the flight mechanics of birds, particularly seagulls. It enriches the advantages of light weight, flexible maneuver and low energy consumption. The flight, however, is affected by elastic vibration and system uncertainty, which can degrade the control performance. To address these issues, this paper proposes a composite learning (CL) control method for the BRFCFW. Through the improved rigid finite element (IRFE) formulation, a dynamic model of the flapping wing system is developed and visualized in MapleSim. Considering non-minimum phase behavior, the system dynamics are transformed into two subsystems through output redefinition: the inner system and the input-output system. For the inner system, we employ the critical gain method to adjust the proportional-derivative (PD) parameters. In the input-output system, the closed-loop control is designed to account for unknown dynamics by utilizing neural modeling error. Stability analysis of the closed-loop system is conducted using the Lyapunov method, and co-simulation is performed with MapleSim and MATLAB/Simulink software. At the end, the feasibility of the recommended strategy is confirmed by comparing it with PD control and neural network (NN) control strategies.