基于动态模糊神经网络的飞行仿真模拟转台控制

给出了一种新型的动态模糊神经网络及其详细的推理过程和参数迭代算法，并应用于高精度飞行仿真模拟转台控制，控制方法结合了神经网络、模糊推理和动态反馈控制的优点，结构简单、收敛速度快、实时性好。实际控制效果表明该控制方法比PID 前馈控制具有更好的控制性能，并且具有更强的抗干扰能力和鲁棒性。

Dynamic Fuzzy-Neural Network Control of Flight Simulator

A novel dynamic fuzzy-neural network (DFNN) is proposed. The DFNN combines the advantage of fuzzy system, neural network and dynamic feedback control. It can not only reduce the time of neural network convergence, but also findsthe optimum rule of fuzzy reasoning. The results of a experiment on flight simulator demonstrate that the proposed controller has a better response than traditional PID forward controller especially and has better robustness against parameter uncertainty and extern disturbance.