Advertisement

Autonomous Inverted Helicopter Flight via Reinforcement Learning

  • Andrew Y. Ng
  • Adam Coates
  • Mark Diel
  • Varun Ganapathi
  • Jamie Schulte
  • Ben Tse
  • Eric Berger
  • Eric Liang
IX. Flying Robots
Part of the Springer Tracts in Advanced Robotics book series (STAR, volume 21)

Abstract

Helicopters have highly stochastic, nonlinear, dynamics, and autonomous helicopter flight is widely regarded to be a challenging control problem. As helicopters are highly unstable at low speeds, it is particularly difficult to design controllers for low speed aerobatic maneuvers. In this paper, we describe a successful application of reinforcement learning to designing a controller for sustained inverted flight on an autonomous helicopter. Using data collected from the helicopter in flight, we began by learning a stochastic, nonlinear model of the helicopter’s dynamics. Then, a reinforcement learning algorithm was applied to automatically learn a controller for autonomous inverted hovering. Finally, the resulting controller was successfully tested on our autonomous helicopter platform.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Andrew Y. Ng
    • 1
  • Adam Coates
    • 1
  • Mark Diel
    • 2
  • Varun Ganapathi
    • 1
  • Jamie Schulte
    • 1
  • Ben Tse
    • 2
  • Eric Berger
    • 1
  • Eric Liang
    • 1
  1. 1.Computer Science Department, Stanford University, Stanford, CA 94305USA
  2. 2.Whirled Air Helicopters, Menlo Park, CA 94025USA

Personalised recommendations