Abstract:

We consider the problem of learning to follow a desired trajectory when given a small number of demonstrations from a sub-optimal expert. We present an algorithm that (i) extracts the---initially unknown---desired trajectory from the sub-optimal expert's demonstrations and (ii) learns a local model suitable for control along the learned trajectory. We apply our algorithm to the problem of autonomous helicopter flight. In all cases, the autonomous helicopter's performance exceeds that of our expert helicopter pilot's demonstrations. Even stronger, our results significantly extend the state-of-the-art in autonomous helicopter aerobatics. In particular, our results include the first autonomous tic-tocs, loops and hurricane, vastly superior performance on previously performed aerobatic maneuvers (such as in-place flips and rolls), and a complete airshow, which requires autonomous transitions between these and various other maneuvers.

Full Paper

Learning for Control from Multiple Demonstrations. Adam Coates, Pieter Abbeel, and Andrew Y. Ng. To appear in ICML, 2008. [ps, pdf]

ICML Joint Paper Session Talk

ICML Joint Session Talk. Helsinki, Finland; July 6, 2008. [VideoLectures.net]
ICML Talk Slides. [PPT]

Videos

The videos below show the results from our algorithm.
  • Recorded demonstrations videos show visualizations of the recorded pilot demonstrations for each maneuver.
  • Aligned demonstrations videos show visualizations of the aligned demonstration trajectories, along with the inferred ideal trajectory (shown as a white helicopter).
  • Flight performance video shows a comparison of the actual helicopter trajectory (black helicopter) along with the desired ideal trajectory (white helicopter), as displayed by our software during an autonomous flight on the real helicopter.
  • Autonomous flight videos show the real autonomous flight of the helicopter.
Airshow
Recorded demonstrations [mov]
Aligned demonstrations [mov]
Flight performance [mov]
Autonomous flight [mp4 (720x405), mp4 (960x540), wmv (960x540)]
Tic Toc
Recorded demonstrations [mov]
Aligned demonstrations [mov]
Flight performance [mov]
Autonomous flight [mp4 (640x360), mp4 (960x540) ]
Flips
Recorded demonstrations [mov]
Aligned demonstrations [mov]
Flight performance [mov]
Autonomous flight [mp4 (640x360), mp4 (960x540) ]
Rolls
Recorded demonstrations [mov]
Aligned demonstrations [mov]
Flight performance [mov]
Autonomous flight [mp4 (640x360), mp4 (960x540) ]