• DocumentCode
    1438639
  • Title

    Avalon

  • Author

    Erckens, Hendrik ; Busser, G.-A. ; Pradalier, Cédric ; Siegwart, Roland Y.

  • Author_Institution
    ETH Zurich, Zurich, Switzerland
  • Volume
    17
  • Issue
    1
  • fYear
    2010
  • fDate
    3/1/2010 12:00:00 AM
  • Firstpage
    45
  • Lastpage
    54
  • Abstract
    This article describes the design and implementation of a navigation and control system for the autonomous sailing vessel Avalon. This boat, designed for participating in the Microtransat, is engineered to autonomously cross the Atlantic Ocean. We present here a special robust mechanical design and the navigation software that will plan an optimal navigation course and efficiently control the boat on this course. The path planner uses an A* algorithm to generate the fastest path to a given destination. It is able to avoid both static and dynamic obstacles. By using a given polar diagram and measured wind data, it takes into account maneuverability constraints of a sailboat. The control system takes care of rudder and sail to follow a given heading while optimizing speed. Additionally, the system is capable of automatically conducting maneuvers such as tack and jibe. At the time of this writing, Avalon and its software systems have been successfully tested more than two weeks in short deployments lasting several hours with winds ranging from zero to 30 kn.
  • Keywords
    boats; marine engineering; mobile robots; navigation; optimisation; path planning; underwater vehicles; vehicle dynamics; velocity control; Avalon; Microtransat; autonomous sailing vessel; boat control; control system; maneuverability constraints; navigation software; navigation system; obstacles avoidance; optimal navigation course; path planner; polar diagram; robust mechanical design; rudder; speed optimization; wind data; Automatic control; Boats; Control systems; Design engineering; Navigation; Oceans; Optimal control; Robust control; Sea measurements; Software systems;
  • fLanguage
    English
  • Journal_Title
    Robotics & Automation Magazine, IEEE
  • Publisher
    ieee
  • ISSN
    1070-9932
  • Type

    jour

  • DOI
    10.1109/MRA.2010.935792
  • Filename
    5430391