Cargando…
A TinyML Deep Learning Approach for Indoor Tracking of Assets †
Positioning systems have gained paramount importance for many different productive sector; however, traditional systems such as Global Positioning System (GPS) have failed to offer accurate and scalable solutions for indoor positioning requirements. Nowadays, alternative solutions such as fingerprin...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9921810/ https://www.ncbi.nlm.nih.gov/pubmed/36772582 http://dx.doi.org/10.3390/s23031542 |
_version_ | 1784887401068363776 |
---|---|
author | Avellaneda, Diego Mendez, Diego Fortino, Giancarlo |
author_facet | Avellaneda, Diego Mendez, Diego Fortino, Giancarlo |
author_sort | Avellaneda, Diego |
collection | PubMed |
description | Positioning systems have gained paramount importance for many different productive sector; however, traditional systems such as Global Positioning System (GPS) have failed to offer accurate and scalable solutions for indoor positioning requirements. Nowadays, alternative solutions such as fingerprinting allow the recognition of the characteristic signature of a location based on RF signal acquisition. In this work, a machine learning (ML) approach has been considered in order to classify the RSSI information acquired by multiple scanning stations from TAG broadcasting messages. TinyML has been considered for this project, as it is a rapidly growing technological paradigm that aims to assist the design and implementation of ML mechanisms in resource-constrained embedded devices. Hence, this paper presents the design, implementation, and deployment of embedded devices capable of communicating and sending information to a central system that determines the location of objects in a defined environment. A neural network (deep learning) is trained and deployed on the edge, allowing the multiple external error factors that affect the accuracy of traditional position estimation algorithms to be considered. Edge Impulse is selected as the main platform for data standardization, pre-processing, model training, evaluation, and deployment. The final deployed system is capable of classifying real data from the installed TAGs, achieving a classification accuracy of [Formula: see text] , which can be increased to [Formula: see text] when a post-processing stage is implemented. |
format | Online Article Text |
id | pubmed-9921810 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-99218102023-02-12 A TinyML Deep Learning Approach for Indoor Tracking of Assets † Avellaneda, Diego Mendez, Diego Fortino, Giancarlo Sensors (Basel) Article Positioning systems have gained paramount importance for many different productive sector; however, traditional systems such as Global Positioning System (GPS) have failed to offer accurate and scalable solutions for indoor positioning requirements. Nowadays, alternative solutions such as fingerprinting allow the recognition of the characteristic signature of a location based on RF signal acquisition. In this work, a machine learning (ML) approach has been considered in order to classify the RSSI information acquired by multiple scanning stations from TAG broadcasting messages. TinyML has been considered for this project, as it is a rapidly growing technological paradigm that aims to assist the design and implementation of ML mechanisms in resource-constrained embedded devices. Hence, this paper presents the design, implementation, and deployment of embedded devices capable of communicating and sending information to a central system that determines the location of objects in a defined environment. A neural network (deep learning) is trained and deployed on the edge, allowing the multiple external error factors that affect the accuracy of traditional position estimation algorithms to be considered. Edge Impulse is selected as the main platform for data standardization, pre-processing, model training, evaluation, and deployment. The final deployed system is capable of classifying real data from the installed TAGs, achieving a classification accuracy of [Formula: see text] , which can be increased to [Formula: see text] when a post-processing stage is implemented. MDPI 2023-01-31 /pmc/articles/PMC9921810/ /pubmed/36772582 http://dx.doi.org/10.3390/s23031542 Text en © 2023 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 Avellaneda, Diego Mendez, Diego Fortino, Giancarlo A TinyML Deep Learning Approach for Indoor Tracking of Assets † |
title | A TinyML Deep Learning Approach for Indoor Tracking of Assets † |
title_full | A TinyML Deep Learning Approach for Indoor Tracking of Assets † |
title_fullStr | A TinyML Deep Learning Approach for Indoor Tracking of Assets † |
title_full_unstemmed | A TinyML Deep Learning Approach for Indoor Tracking of Assets † |
title_short | A TinyML Deep Learning Approach for Indoor Tracking of Assets † |
title_sort | tinyml deep learning approach for indoor tracking of assets † |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9921810/ https://www.ncbi.nlm.nih.gov/pubmed/36772582 http://dx.doi.org/10.3390/s23031542 |
work_keys_str_mv | AT avellanedadiego atinymldeeplearningapproachforindoortrackingofassets AT mendezdiego atinymldeeplearningapproachforindoortrackingofassets AT fortinogiancarlo atinymldeeplearningapproachforindoortrackingofassets AT avellanedadiego tinymldeeplearningapproachforindoortrackingofassets AT mendezdiego tinymldeeplearningapproachforindoortrackingofassets AT fortinogiancarlo tinymldeeplearningapproachforindoortrackingofassets |