航空学报 > 1995, Vol. 16 Issue (4): 48-53

基于反馈误差学习的神经网络控制

任雪梅, 高为炳   

  1. 北京航空航天大学第七研究室
  • 收稿日期:1994-01-27 修回日期:1994-06-06 出版日期:1995-08-25 发布日期:1995-08-25

NEURAL NETWORK CONTROL USING FEEDBACK ERROR LEARNING

Ren Xuemei,Gao Weibing   

  1. 7th Research Division of Beijing University of Aeronautics and Astronautics,Beijing,100083
  • Received:1994-01-27 Revised:1994-06-06 Online:1995-08-25 Published:1995-08-25

摘要: 研究了应用神经网络和PD反馈控制实现非线性系统的自适应跟踪问题。PD反馈控制器不但保证闭环系统的稳定性,同时其输出又作为训练神网络的参考信号。证明了通过选择适当的初始加权和加权的调节速率可以实现非线性系统的固定点跟踪。

Abstract: Neural networks and PD feedback controllers for adaptive tracking in nonlinear systems are presented. The PD feedback controller is used as a general feedback controller to guarantee global stability,and its output is fed to the neural network as the reference signal. The asymptotic set-point tracking of the system can be guaranteed by selecting the appropriate initial weights and the updating rate of the weights of the networks.

中图分类号: