【摘要】 基于对微型扑翼飞行器模型的认识，建立了数学模型描述微型扑翼飞行器在运行过程中产生的气动力和气动力矩，并建立了微型扑翼飞行器的运动模型。基于RBF神经网络的万能逼近性实现对FWMAV中运动模型中不确定函数的实时估计，并证明上述神经网络可以稳定跟踪运动模型中的不确定函数，并根据估计的函数结果用于PD控制实现微型扑翼飞行器对设定姿态角的跟踪。最终，设计了测试方案并通过MATLAB/Simulink环境仿真的方式验证了上述方法的有效性和可操作性。更多还原

【Abstract】 Based on the understanding of the miniature flapping wing vehicle(FWMAV) model, a mathematical model is established to describe the aerodynamic forces and aerodynamic moments generated by the FWMAV during operation, and the motion model of the FWMAV is established. The real-time estimation of the uncertain function in the motion model in FWMAV is achieved based on the universal approximation of the RBF neural network, and it is proved that the neural network can stably track the uncertain function in the motion model and be used for the PD control to achieve the tracking of the miniature flapping wing vehicle to the set attitude angle based on the results of the estimated function. Finally, a test scheme is designed and the effectiveness and operability of the method is verified by means of simulation in MATLAB/Simulink environment.更多还原