Enhancing space sensor resilience with transfer learning in data-scarce scenarios

In Mars exploration missions, harsh environmental conditions, such as those generated by dust devils, can damage sensing systems. Soft sensors offer a promising solution in such scenarios, but their implementation is challenging when data is scarce. This paper explores the use of transfer learning (...

Full description

Saved in:
Bibliographic Details
Main Authors: Dileep Kumar, Manuel Domínguez-Pumar, Beatriz Otero-Calviño, Joan Pons-Nin, Josefina Torres, Mercedes Marín, Javier Gómez-Elvira, Luis Mora, Sara Navarro, Jose Rodríguez-Manfredi
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:Machine Learning: Science and Technology
Subjects:
Online Access:https://doi.org/10.1088/2632-2153/adfd38
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849222177740554240
author Dileep Kumar
Manuel Domínguez-Pumar
Beatriz Otero-Calviño
Joan Pons-Nin
Josefina Torres
Mercedes Marín
Javier Gómez-Elvira
Luis Mora
Sara Navarro
Jose Rodríguez-Manfredi
author_facet Dileep Kumar
Manuel Domínguez-Pumar
Beatriz Otero-Calviño
Joan Pons-Nin
Josefina Torres
Mercedes Marín
Javier Gómez-Elvira
Luis Mora
Sara Navarro
Jose Rodríguez-Manfredi
author_sort Dileep Kumar
collection DOAJ
description In Mars exploration missions, harsh environmental conditions, such as those generated by dust devils, can damage sensing systems. Soft sensors offer a promising solution in such scenarios, but their implementation is challenging when data is scarce. This paper explores the use of transfer learning (TL) to enhance sensor resilience, specifically addressing the issue of limited data availability. First, pre-trained models are developed using wind sensor data from the Temperature and Wind Sensors for the TWINS instrument (NASA InSight Mission), serving as the source domain for this study. These models account for the malfunction of a single wind sensing board or transducer. The pre-trained models are then adapted using TL to the Mars Environmental Dynamics Analyzer (MEDA) wind sensor (NASA Perseverance Mission), which shares common sensing principles, and which suffered a malfunction during the mission. Hyperparameter tuning further improves the performance of the TL models, yielding better results than models trained solely on a small MEDA dataset. The results demonstrate the effectiveness of the TL-based approach in recovering variables from the MEDA wind sensor despite partial failures and data limitations. Overall, the TL-based method improves performance by 10.21%–22.04% compared to models trained exclusively on the limited MEDA dataset.
format Article
id doaj-art-96bdfb96dbad47bd8c909f18e7ac72ec
institution Kabale University
issn 2632-2153
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series Machine Learning: Science and Technology
spelling doaj-art-96bdfb96dbad47bd8c909f18e7ac72ec2025-08-26T07:16:50ZengIOP PublishingMachine Learning: Science and Technology2632-21532025-01-016303LT0110.1088/2632-2153/adfd38Enhancing space sensor resilience with transfer learning in data-scarce scenariosDileep Kumar0https://orcid.org/0000-0002-6211-1078Manuel Domínguez-Pumar1https://orcid.org/0000-0001-5439-7953Beatriz Otero-Calviño2https://orcid.org/0000-0002-9194-559XJoan Pons-Nin3https://orcid.org/0000-0002-0356-5678Josefina Torres4https://orcid.org/0000-0003-1035-6740Mercedes Marín5https://orcid.org/0000-0003-2328-1303Javier Gómez-Elvira6https://orcid.org/0000-0002-9068-9846Luis Mora7https://orcid.org/0000-0002-8209-1190Sara Navarro8https://orcid.org/0000-0001-8606-7799Jose Rodríguez-Manfredi9https://orcid.org/0000-0003-0461-9815Universitat Politècnica de Catalunya (UPC) , Barcelona, SpainUniversitat Politècnica de Catalunya (UPC) , Barcelona, SpainDepartment of Computer Architecture, Universitat Politècnica de Catalunya (UPC) , Barcelona, SpainUniversitat Politècnica de Catalunya (UPC) , Barcelona, SpainCentro de Astrobiología (INTA-CSIC) , Madrid, SpainCentro de Astrobiología (INTA-CSIC) , Madrid, SpainCentro de Astrobiología (INTA-CSIC) , Madrid, SpainCentro de Astrobiología (INTA-CSIC) , Madrid, SpainCentro de Astrobiología (INTA-CSIC) , Madrid, SpainCentro de Astrobiología (INTA-CSIC) , Madrid, SpainIn Mars exploration missions, harsh environmental conditions, such as those generated by dust devils, can damage sensing systems. Soft sensors offer a promising solution in such scenarios, but their implementation is challenging when data is scarce. This paper explores the use of transfer learning (TL) to enhance sensor resilience, specifically addressing the issue of limited data availability. First, pre-trained models are developed using wind sensor data from the Temperature and Wind Sensors for the TWINS instrument (NASA InSight Mission), serving as the source domain for this study. These models account for the malfunction of a single wind sensing board or transducer. The pre-trained models are then adapted using TL to the Mars Environmental Dynamics Analyzer (MEDA) wind sensor (NASA Perseverance Mission), which shares common sensing principles, and which suffered a malfunction during the mission. Hyperparameter tuning further improves the performance of the TL models, yielding better results than models trained solely on a small MEDA dataset. The results demonstrate the effectiveness of the TL-based approach in recovering variables from the MEDA wind sensor despite partial failures and data limitations. Overall, the TL-based method improves performance by 10.21%–22.04% compared to models trained exclusively on the limited MEDA dataset.https://doi.org/10.1088/2632-2153/adfd38Martian wind sensorssoft sensorsdeep learningtransfer learningsmall dataset
spellingShingle Dileep Kumar
Manuel Domínguez-Pumar
Beatriz Otero-Calviño
Joan Pons-Nin
Josefina Torres
Mercedes Marín
Javier Gómez-Elvira
Luis Mora
Sara Navarro
Jose Rodríguez-Manfredi
Enhancing space sensor resilience with transfer learning in data-scarce scenarios
Machine Learning: Science and Technology
Martian wind sensors
soft sensors
deep learning
transfer learning
small dataset
title Enhancing space sensor resilience with transfer learning in data-scarce scenarios
title_full Enhancing space sensor resilience with transfer learning in data-scarce scenarios
title_fullStr Enhancing space sensor resilience with transfer learning in data-scarce scenarios
title_full_unstemmed Enhancing space sensor resilience with transfer learning in data-scarce scenarios
title_short Enhancing space sensor resilience with transfer learning in data-scarce scenarios
title_sort enhancing space sensor resilience with transfer learning in data scarce scenarios
topic Martian wind sensors
soft sensors
deep learning
transfer learning
small dataset
url https://doi.org/10.1088/2632-2153/adfd38
work_keys_str_mv AT dileepkumar enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT manueldominguezpumar enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT beatrizoterocalvino enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT joanponsnin enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT josefinatorres enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT mercedesmarin enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT javiergomezelvira enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT luismora enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT saranavarro enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios
AT joserodriguezmanfredi enhancingspacesensorresiliencewithtransferlearningindatascarcescenarios