航空学报 > 2002, Vol. 23 Issue (6): 530-533

基于径向基神经网络的不确定非线性系统的鲁棒自适应控制

武国庆, 姜长生, 张锐, 卢伟健   

  1. 南京航空航天大学自动化学院 江苏南京 210016
  • 收稿日期:2001-09-03 修回日期:2002-08-20 出版日期:2002-12-25 发布日期:2002-12-25

ROBUST ADAPTIVE CONTROL OF UNCERTAIN NONLINEAR SYSTEMS BASED ON RBF NEURAL NETWORK

WU Guo-qing,JIANG Chang-sheng,ZHANG Rui,LU Wei-jian   

  1. Automation College, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Received:2001-09-03 Revised:2002-08-20 Online:2002-12-25 Published:2002-12-25

摘要:

针对能够采用仿射非线性表示的含有不确定动态的非线性系统 ,提出了一种鲁棒自适应控制方法 ,该方法根据离线辨识出的受控对象的已知部分 ,采用神经网络在线辨识其未知部分 ,并针对辨识得到神经网络模型采用反馈线性化方法设计出自适应控制器 ,同时引入滑模控制方法以增强控制系统的鲁棒性 ,从而实现鲁棒自适应控制。通过对具有未建模动态的非线性直升机空气动力学模型 ,设计了总距通道系统。仿真表明该方法是有效的。

关键词: 非线性控制, 不确定系统, 神经网络, 滑模控制

Abstract:

A new approach for the robust control of uncertain affined nonlinear systems is presented.Based on the certain part of the plant's model identified off line,neural networks are used in identifying the uncertainty on line.Then the robust adaptive control is realized by applying feedback linearizing with the neural networks model and the robustness of the control system is enhanced by sliding model control.Finally,the approach is applied to a collective pitch angle channel for a simplified nonlinear helicopter aerodynamic model with input unmodeled dynamics.Simulation results demonstrate the validity of the approach.

Key words: nonlinear control, uncertainty system, neural network, sliding model control