论文信息 - Autonomous Helicopter Flight via Reinforcement Learning

Autonomous Helicopter Flight via Reinforcement Learning

Autonomous helicopter flight represents a challenging control problem, with complex, noisy, dynamics. In this paper, we describe a successful application of reinforcement learning to autonomous helicopter flight. We first fit a stochastic, nonlinear model of the helicopter dynamics. We then use the model to learn to hover in place, and to fly a number of maneuvers taken from an RC helicopter competition.

[1] J. Kiefer,et al. Stochastic Estimation of the Maximum of a Regression Function , 1952 .

[2] W. Cleveland. Robust Locally Weighted Regression and Smoothing Scatterplots , 1979 .

[3] Gene F. Franklin,et al. Feedback Control of Dynamic Systems , 1986 .

[4] Keith Glover,et al. μ-analysis and synthesis toolbox: for use with Matlab, user’s guide version 3 , 1998 .

[5] Andrew Y. Ng,et al. Policy Invariance Under Reward Transformations: Theory and Application to Reward Shaping , 1999, ICML.

[6] J. Gordon Leishman,et al. Principles of Helicopter Aerodynamics , 2000 .

[7] S. Shankar Sastry,et al. HIERARCHICAL CONTROL SYSTEM SYNTHESIS FOR ROTORCRAFT-BASED UNMANNED AERIAL VEHICLES , 2000 .

[8] Michael I. Jordan,et al. PEGASUS: A policy search method for large MDPs and POMDPs , 2000, UAI.

[9] Jeff G. Schneider,et al. Autonomous helicopter control using reinforcement learning policy search methods , 2001, Proceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (Cat. No.01CH37164).

[10] Andrew Y. Ng,et al. Shaping and policy search in reinforcement learning , 2003 .

[11] Andrew W. Moore,et al. Locally Weighted Learning , 1997, Artificial Intelligence Review.