Cargando…
A Simulator and First Reinforcement Learning Results for Underwater Mapping
Underwater mapping with mobile robots has a wide range of applications, and good models are lacking for key parts of the problem, such as sensor behavior. The specific focus here is the huge environmental problem of underwater litter, in the context of the Horizon 2020 SeaClear project, where a team...
Autores principales: | , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9322081/ https://www.ncbi.nlm.nih.gov/pubmed/35891061 http://dx.doi.org/10.3390/s22145384 |
_version_ | 1784756209399627776 |
---|---|
author | Rosynski, Matthias Buşoniu, Lucian |
author_facet | Rosynski, Matthias Buşoniu, Lucian |
author_sort | Rosynski, Matthias |
collection | PubMed |
description | Underwater mapping with mobile robots has a wide range of applications, and good models are lacking for key parts of the problem, such as sensor behavior. The specific focus here is the huge environmental problem of underwater litter, in the context of the Horizon 2020 SeaClear project, where a team of robots is being developed to map and collect such litter. No reinforcement-learning solution to underwater mapping has been proposed thus far, even though the framework is well suited for robot control in unknown settings. As a key contribution, this paper therefore makes a first attempt to apply deep reinforcement learning (DRL) to this problem by exploiting two state-of-the-art algorithms and making a number of mapping-specific improvements. Since DRL often requires millions of samples to work, a fast simulator is required, and another key contribution is to develop such a simulator from scratch for mapping seafloor objects with an underwater vehicle possessing a sonar-like sensor. Extensive numerical experiments on a range of algorithm variants show that the best DRL method collects litter significantly faster than a baseline lawn mower trajectory. |
format | Online Article Text |
id | pubmed-9322081 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-93220812022-07-27 A Simulator and First Reinforcement Learning Results for Underwater Mapping Rosynski, Matthias Buşoniu, Lucian Sensors (Basel) Article Underwater mapping with mobile robots has a wide range of applications, and good models are lacking for key parts of the problem, such as sensor behavior. The specific focus here is the huge environmental problem of underwater litter, in the context of the Horizon 2020 SeaClear project, where a team of robots is being developed to map and collect such litter. No reinforcement-learning solution to underwater mapping has been proposed thus far, even though the framework is well suited for robot control in unknown settings. As a key contribution, this paper therefore makes a first attempt to apply deep reinforcement learning (DRL) to this problem by exploiting two state-of-the-art algorithms and making a number of mapping-specific improvements. Since DRL often requires millions of samples to work, a fast simulator is required, and another key contribution is to develop such a simulator from scratch for mapping seafloor objects with an underwater vehicle possessing a sonar-like sensor. Extensive numerical experiments on a range of algorithm variants show that the best DRL method collects litter significantly faster than a baseline lawn mower trajectory. MDPI 2022-07-19 /pmc/articles/PMC9322081/ /pubmed/35891061 http://dx.doi.org/10.3390/s22145384 Text en © 2022 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 Rosynski, Matthias Buşoniu, Lucian A Simulator and First Reinforcement Learning Results for Underwater Mapping |
title | A Simulator and First Reinforcement Learning Results for Underwater Mapping |
title_full | A Simulator and First Reinforcement Learning Results for Underwater Mapping |
title_fullStr | A Simulator and First Reinforcement Learning Results for Underwater Mapping |
title_full_unstemmed | A Simulator and First Reinforcement Learning Results for Underwater Mapping |
title_short | A Simulator and First Reinforcement Learning Results for Underwater Mapping |
title_sort | simulator and first reinforcement learning results for underwater mapping |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9322081/ https://www.ncbi.nlm.nih.gov/pubmed/35891061 http://dx.doi.org/10.3390/s22145384 |
work_keys_str_mv | AT rosynskimatthias asimulatorandfirstreinforcementlearningresultsforunderwatermapping AT busoniulucian asimulatorandfirstreinforcementlearningresultsforunderwatermapping AT rosynskimatthias simulatorandfirstreinforcementlearningresultsforunderwatermapping AT busoniulucian simulatorandfirstreinforcementlearningresultsforunderwatermapping |