Cargando…

Deep Learning for Efficient and Optimal Motion Planning for AUVs with Disturbances

We use the recent advances in Deep Learning to solve an underwater motion planning problem by making use of optimal control tools—namely, we propose using the Deep Galerkin Method (DGM) to approximate the Hamilton–Jacobi–Bellman PDE that can be used to solve continuous time and state optimal control...

Descripción completa

Detalles Bibliográficos
Autores principales: Parras, Juan, Apellániz, Patricia A., Zazo, Santiago
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8347268/
https://www.ncbi.nlm.nih.gov/pubmed/34372249
http://dx.doi.org/10.3390/s21155011
_version_ 1783735045547098112
author Parras, Juan
Apellániz, Patricia A.
Zazo, Santiago
author_facet Parras, Juan
Apellániz, Patricia A.
Zazo, Santiago
author_sort Parras, Juan
collection PubMed
description We use the recent advances in Deep Learning to solve an underwater motion planning problem by making use of optimal control tools—namely, we propose using the Deep Galerkin Method (DGM) to approximate the Hamilton–Jacobi–Bellman PDE that can be used to solve continuous time and state optimal control problems. In order to make our approach more realistic, we consider that there are disturbances in the underwater medium that affect the trajectory of the autonomous vehicle. After adapting DGM by making use of a surrogate approach, our results show that our method is able to efficiently solve the proposed problem, providing large improvements over a baseline control in terms of costs, especially in the case in which the disturbances effects are more significant.
format Online
Article
Text
id pubmed-8347268
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-83472682021-08-08 Deep Learning for Efficient and Optimal Motion Planning for AUVs with Disturbances Parras, Juan Apellániz, Patricia A. Zazo, Santiago Sensors (Basel) Article We use the recent advances in Deep Learning to solve an underwater motion planning problem by making use of optimal control tools—namely, we propose using the Deep Galerkin Method (DGM) to approximate the Hamilton–Jacobi–Bellman PDE that can be used to solve continuous time and state optimal control problems. In order to make our approach more realistic, we consider that there are disturbances in the underwater medium that affect the trajectory of the autonomous vehicle. After adapting DGM by making use of a surrogate approach, our results show that our method is able to efficiently solve the proposed problem, providing large improvements over a baseline control in terms of costs, especially in the case in which the disturbances effects are more significant. MDPI 2021-07-23 /pmc/articles/PMC8347268/ /pubmed/34372249 http://dx.doi.org/10.3390/s21155011 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Parras, Juan
Apellániz, Patricia A.
Zazo, Santiago
Deep Learning for Efficient and Optimal Motion Planning for AUVs with Disturbances
title Deep Learning for Efficient and Optimal Motion Planning for AUVs with Disturbances
title_full Deep Learning for Efficient and Optimal Motion Planning for AUVs with Disturbances
title_fullStr Deep Learning for Efficient and Optimal Motion Planning for AUVs with Disturbances
title_full_unstemmed Deep Learning for Efficient and Optimal Motion Planning for AUVs with Disturbances
title_short Deep Learning for Efficient and Optimal Motion Planning for AUVs with Disturbances
title_sort deep learning for efficient and optimal motion planning for auvs with disturbances
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8347268/
https://www.ncbi.nlm.nih.gov/pubmed/34372249
http://dx.doi.org/10.3390/s21155011
work_keys_str_mv AT parrasjuan deeplearningforefficientandoptimalmotionplanningforauvswithdisturbances
AT apellanizpatriciaa deeplearningforefficientandoptimalmotionplanningforauvswithdisturbances
AT zazosantiago deeplearningforefficientandoptimalmotionplanningforauvswithdisturbances