华南理工大学学报(自然科学版) ›› 2008, Vol. 36 ›› Issue (6): 90-94.

• 交通运输工程 • 上一篇    下一篇

基于视觉和后推方法的智能车轨迹跟踪控制

卞建勇徐建闽1  胡跃明2   

  1. 1. 华南理工大学 土木与交通学院, 广东 广州 510640;2. 华南理工大学 自动化科学与工程学院, 广东 广州 510640
  • 收稿日期:2007-04-26 修回日期:2007-09-19 出版日期:2008-06-25 发布日期:2008-06-25
  • 通信作者: 卞建勇(1980-),男,博士生,主要从事非线性控制、智能控制、图像处理与模式识别及智能交通研究. E-mail:bjyong977@126.com
  • 作者简介:卞建勇(1980-),男,博士生,主要从事非线性控制、智能控制、图像处理与模式识别及智能交通研究.
  • 基金资助:

    国家“863”计划项目(2006AA11Z211);广东省工业科技攻关项目(B04B2051550)

Trajectory Tracking Control of Intelligent Vehicle Based on Vision and Backstepping Method

Bian Jian-yong1  Xu Jian-min1  Hu Yue-ming2   

  1. 1. School of Civil Engineering and Transportation, South China University of Technology, Guangzhou 510640, Guangdong, China; 2. School of Automation Science and Engineering, South China University of Technology, Guangzhou 510640, Guangdong, China
  • Received:2007-04-26 Revised:2007-09-19 Online:2008-06-25 Published:2008-06-25
  • Contact: 卞建勇(1980-),男,博士生,主要从事非线性控制、智能控制、图像处理与模式识别及智能交通研究. E-mail:bjyong977@126.com
  • About author:卞建勇(1980-),男,博士生,主要从事非线性控制、智能控制、图像处理与模式识别及智能交通研究.
  • Supported by:

    国家“863”计划项目(2006AA11Z211);广东省工业科技攻关项目(B04B2051550)

摘要: 通过基于视觉的车道标志线识别系统建立智能车的期望跟踪轨迹,并将智能车运动学模型转换为链式系统模型,同时利用后推方法设计控制律,克服了采用动态反馈线性化方法设计的控制器维数较高以及滑模变结构控制器易出现高频抖振的缺点.仿真结果表明:该方法具有较好的轨迹跟踪控制效果和全局稳定性.

关键词: 车道标志线识别, 智能车, 链式系统, 视觉, 后推方法, 轨迹跟踪

Abstract:

In this paper, a vision-based lane marking detection system is adopted to establish the anticipant tracking trajectory of an intelligent vehicle and to transform the kinematic model of the intelligent vehicle into a chained one. Moreover, the backstepping method is employed to design the control rule, thus overcoming not only the defect of high dimension arising from the dynamic feedback linearization controller but also the problem of high-frequency shake existing in the sliding-mode controller. Simulated results indicate that the proposed method is of good trajectory tracking control ability and global stability.

Key words: lane marking detection, intelligent vehicle, chained system, vision, backstepping method, trajectory tracking