航空学报 > 2004, Vol. 25 Issue (2): 148-152

基于物理模型的拟人智能控制

李诚, 张明廉   

  1. 北京航空航天大学自动化学院自动控制系 北京 100083
  • 收稿日期:2003-02-15 修回日期:2003-07-07 出版日期:2004-04-25 发布日期:2004-04-25

Human-Imitating Intelligent Control Based on Physical Model

LI Cheng, ZHANG Ming-lian   

  1. Department of Automatic Control in School of Automation Science and Electrical Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100083, China
  • Received:2003-02-15 Revised:2003-07-07 Online:2004-04-25 Published:2004-04-25

摘要: 回顾了自动控制理论与智能控制的过去和现在,讨论了它们存在的问题,提出基于物理模型的拟人智能控制。通过介绍一个成功的实例"平面运动单级倒立摆"的控制来阐明拟人智能控制的机理以及设计控制律的流程。对于复杂的非线性被控对象,只要知道其物理模型,应用拟人智能控制方法可以得到很好的结果。

关键词: 智能控制, 归约, 模糊控制, 人工神经网络

Abstract: Based on the review of traditional control theories and some intelligent control theories,their limitations are discussed,then the human-imitating intelligent control method is presented which is based on the physical model.The mechanism of this method is explained and the procedure of designing the human-imitating control law is introduced through its successful application to stabilizing the two-degree-of-freedom single inverted pendulum.As for the complicated nonlinear objects,applying this human-imitating intelligent control method can achieve satisfactory results as long as the object's physical model is available.

Key words: intelligent control, reduction, fuzzy control, artificial neural network