Intelligent Flapping Wing Control

Reinforcement Learning for the DelFly

Master thesis (2017)

Authors

M.W. Goedhart Aerospace Engineering

Contributors

E. van Kampen (supervisor 1)
S.F. Armanini (supervisor 1)
C.C. de Visser (supervisor 1)
Q. P. Chu (supervisor 2)
Faculty
Aerospace Engineering
Copyright: © 2017 Menno Goedhart
Reinforcement Learning Machine Learning Classification Flapping Wing Policy Gradient DelFly
More Info
expand_more

Published Date

23-06-2017

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Aerospace Engineering

Abstract

Flight control of the DelFly is challenging, because of its complex dynamics and variability due to manufacturing inconsistencies. Machine Learning algorithms can be used to tackle these challenges. A Policy Gradient algorithm is used to tune the gains of a Proportional-Integral controller using Reinforcement Learning. Furthermore, a novel Classification Algorithm for Machine Learning control (CAML) is presented, which uses model identification and a neural network classifier to select from several predefined gain sets. The algorithms show comparable performance when considering variability only, but the Policy Gradient algorithm is more robust to noise, disturbances, nonlinearities and flapping motion.