Publication Details

Dynamic Soaring in Uncertain Wind Conditions: Polynomial Chaos Expansion Approach

NOVÁK, J.; CHUDÝ, P. Dynamic Soaring in Uncertain Wind Conditions: Polynomial Chaos Expansion Approach. In Machine Learning, Optimization, and Data Science. Lecture Notes in Computer Science. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Grasmere: Springer Nature Switzerland AG, 2024. p. 104-115. ISBN: 978-3-031-53968-8. ISSN: 0302-9743.
Czech title
Dynamické plachtění v nejistých větrných podmínkách: přístup pomocí polynomiálního chaosu
Type
conference paper
Language
English
Authors
Keywords

Polynomial Chaos Expansion, Surrogate Modeling,  Dynamic Soaring, Optimal
Control

Abstract

Dynamic soaring refers to a flight technique used primarily by large seabirds to
extract energy from the wind shear layers formed above ocean surface. A small
Unmanned Aerial Vehicle (UAV) capable of efficient dynamic soaring maneuvers can
enable long endurance missions in context of patrol or increased flight range. To
realize autonomous energy-saving patterns by a UAV, a real-time trajectory
generation for a dynamic soaring maneuver accounting for varying external
conditions has to be performed. The design of the flight trajectory is formulated
as an Optimal Control Problem (OCP) and solved within direct collocation based
optimization. A surrogate model of the optimal traveling cycle capturing wind
profile uncertainties is constructed using Polynomial Chaos Expansion (PCE). The
unknown wind profile parameters are estimated from observed trajectory by means
of a Genetic Algorithm (GA). The PCE surrogate model is subsequently utilized to
update the optimal trajectory using the estimated wind profile parameters.

Published
2024
Pages
104–115
Journal
Lecture Notes in Computer Science, no. 14505, ISSN 0302-9743
Proceedings
Machine Learning, Optimization, and Data Science
Series
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Conference
The 9th International Conference on Machine Learning, Optimization, and Data Science (LOD), Grasmere, GB
ISBN
978-3-031-53968-8
Publisher
Springer Nature Switzerland AG
Place
Grasmere
DOI
EID Scopus
BibTeX
@inproceedings{BUT185184,
  author="Jiří {Novák} and Peter {Chudý}",
  title="Dynamic Soaring in Uncertain Wind Conditions: Polynomial Chaos Expansion Approach",
  booktitle="Machine Learning, Optimization, and Data Science",
  year="2024",
  series="Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
  journal="Lecture Notes in Computer Science",
  number="14505",
  pages="104--115",
  publisher="Springer Nature Switzerland AG",
  address="Grasmere",
  doi="10.1007/978-3-031-53969-5\{_}9",
  isbn="978-3-031-53968-8",
  issn="0302-9743"
}
Back to top