Multiple Look-Ahead Distance Scheme for Lateral Control of Vision-Based Vehicles

In order to realize lateral control of vehicles on an automated highway system (AHS), it is necessary to measure and/or predict vehicle deviation from the path to be followed. An AHS requires a vehicle to be equipped with vision systems and range finders for it to drive intelligently. Vehicles equipped with vision systems can drive to any location regardless of infrastructure limitations. Several methods have already been proposed to solve the automated line-following control problem. In the steering angle of a vehicle is determined based on linear feedback in terms of the vehicle’s heading angle and deviation from the reference line at a look-ahead point; however, it is difficult to implement this method in vision-based vehicles because the vehicle heading determined using a vision system may be sensitive to noise. In a variable multiple look-ahead distance scheme is proposed; among the multiple look-ahead points, one is used for curvature detection and another is used for error detection. In this study, a much simpler multiple look-ahead distance scheme is proposed; in particular, simple linear feedback control of vehicle deviations at multiple look-ahead points is proposed. Simulations are conducted for designing the optimal combination of feedback gains. The designed feedback gains are implemented on an experimental vehicle, and the usefulness of the proposed method is verified.

  • Supplemental Notes:
    • Abstract reprinted with permission from Intelligent Transportation Society of America.
  • Corporate Authors:

    ITS America

    1100 17th Street, NW, 12th Floor
    Washington, DC  United States  20036
  • Authors:
    • Huang, Jingtao
    • Konaka, Eiji
  • Conference:
  • Publication Date: 2012


  • English

Media Info

  • Media Type: Digital/other
  • Features: CD-ROM; Figures; Photos; References; Tables;
  • Pagination: 7p
  • Monograph Title: 19th ITS World Congress, Vienna, Austria, 22 to 26 October 2012

Subject/Index Terms

Filing Info

  • Accession Number: 01499122
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Nov 12 2013 8:58AM