计算机应用 ›› 2013, Vol. 33 ›› Issue (03): 858-861.DOI: 10.3724/SP.J.1087.2013.00858

• 典型应用 • 上一篇    下一篇

四旋翼无人飞行器混合控制系统研究

夏国清1,廖粤峰1*,王璐2   

  1. 1.哈尔滨工程大学 自动化学院,哈尔滨 150001;
    2.上海交通大学 电子信息与电气工程学院,上海 200240
  • 收稿日期:2012-09-14 修回日期:2012-10-26 出版日期:2013-03-01 发布日期:2013-03-01
  • 通讯作者: 廖粤峰(夏国清 转)
  • 作者简介:夏国清(1962-),男,黑龙江哈尔滨人,教授,博士生导师,博士,主要研究方向:船舶动力定位; 廖粤峰(1988-),男(满族),广东阳春人,硕士研究生,主要研究方向:飞行器控制; 王璐(1987-),男,河南洛阳人,博士研究生,主要研究方向:非线性滤波、非线性控制。

Research on hybrid control system of quadrotor UAV

XIA Guoqing1, LIAO Yuefeng1*, WANG Lu2   

  1. 1.Automation College, Harbin Engineering University, Harbin Heilongjiang 150001,China;
    2.College of Electronic Information and Electric Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
  • Received:2012-09-14 Revised:2012-10-26 Online:2013-03-01 Published:2013-03-01
  • Contact: LIAO yuefeng

摘要: 针对四旋翼无人飞行器质量未知情况下的垂直起降控制问题,提出一种基于状态反馈和神经网络自适应的混合控制方法。该方法通过一个状态反馈控制器实现飞行器的水平位置和航向控制,考虑到飞行器负载的未知特性,通过径向基函数(RBF)神经网络对飞行器质量进行估计,从而实现对高度的精确控制。仿真分析及验证表明,所提出的控制方法能够有效实现飞行器高度的精确控制,并能够在线估计出飞行器质量参数。

关键词: 四旋翼无人飞行器, 混合控制, 神经网络, 自适应控制, 径向基函数

Abstract: A hybrid control method based on state feedback and adaptive neural network was proposed, which considered the taking off and landing control problem under unknown mass of the Unmanned Aerial Vehicle (UAV). A state feedback controller was designed to realize the horizontal position and heading control. The accurate control of height was archived considering the vehicle's unknown load through the Radial Basis Function (RBF) neural network. The simulation analysis and experiments illustrate that the proposed control method can effectively realize the accurate control of height, and can be able to online estimate aircraft quality parameters.

Key words: quadrotor Unmanned Aerial Vehicle (UAV), hybrid control, neural network, adaptive control, Radial Basis Function (RBF)

中图分类号: