Autonomous Helicopter Aerobatics through Apprenticeship Learning

TLDR

Autonomous helicopter flight is a highly challenging control problem, yet human experts can reliably perform a wide range of aerobatic maneuvers at the edge of the helicopter’s capabilities. We present apprenticeship learning algorithms that leverage expert demonstrations to efficiently learn good controllers for tasks demonstrated by an expert. The approach uses apprenticeship learning to infer control policies from expert demonstrations. The learned controllers enable autonomous execution of a wide range of aerobatic maneuvers—including flips, rolls, loops, hurricanes, auto‑rotation landings, chaos, and tic‑tocs—and complete airshows, performing as well as or better than the expert pilot.

Abstract

Autonomous helicopter flight is widely regarded to be a highly challenging control problem. Despite this fact, human experts can reliably fly helicopters through a wide range of maneuvers, including aerobatic maneuvers at the edge of the helicopter’s capabilities. We present apprenticeship learning algorithms, which leverage expert demonstrations to efficiently learn good controllers for tasks being demonstrated by an expert. These apprenticeship learning algorithms have enabled us to significantly extend the state of the art in autonomous helicopter aerobatics. Our experimental results include the first autonomous execution of a wide range of maneuvers, including but not limited to in-place flips, in-place rolls, loops and hurricanes, and even auto-rotation landings, chaos and tic-tocs, which only exceptional human pilots can perform. Our results also include complete airshows, which require autonomous transitions between many of these maneuvers. Our controllers perform as well as, and often even better than, our expert pilot.