摘要
讨论了一种基于神经网络动态逆的直接自适应控制方法 ,并应用于超机动飞机的飞行控制中。基本控制律采用非线性动态逆方法进行设计 ,对由于模型不准确导致的逆误差采用单隐层神经网络进行在线补偿。仿真结果表明 ,神经网络通过补偿由于模型不准确引起的逆误差 ,弥补了非线性动态逆要求精确数学模型的缺点 ,提高了整个控制系统的鲁棒性 ,而且可以大大简化动态逆控制律的设计。
An adaptive controller with a neural network compensator is designed and applied in the control of a super-maneuvering aircraft. The base control law is designed by nonlinear dynamic inversion method, and single hidden layer (SHL) neural networks are used to compensate the inversion error induced by the inaccuracy of the system model. Simulational results show that the limitation of the accurate mathematical model used in dynamic inversion method can be released through adaptively canceling inversion error in neural networks and the robustness of the control system is improved. Besides, the design of dynamic inversion control law can be simplified by the online neural networks.
出处
《南京航空航天大学学报》
EI
CAS
CSCD
北大核心
2003年第1期86-90,共5页
Journal of Nanjing University of Aeronautics & Astronautics
关键词
自适应控制
动态逆
神经网络
飞行控制
超机动
adaptive control
dynamic inversion
neural network
flight control
super maneuver
