Autonomous Underwater Vehicles (AUVs) have proven to be a precious resource for oceans preservation thanks to their ability to accomplish survey missions on wide areas without the need of human supervision. This also includes the detection and mapping of underwater gas emissions, whether these are due to damaged offshore structures or naturally released from the seafloor. Indeed, submarine seepage have severe repercussions on the surrounding marine habitat and contribute to greenhouse gas emissions, hence the need to identify and monitor them. This work focuses on the online Automatic Target Recognition (ATR) and geolocalisation of underwater gas leakages utilising an AUV equipped with a Forward Looking Sonar (FLS). With the aim to accurately position the seeps, it was necessary to address the problem of navigating the vehicle in areas with the presence of gas bubbles. In fact, bubbles adversely affect acoustic sensors typically exploited by underwater robots for navigation. The paper investigates the effects that seepage have on the performance of navigation solutions based on two different acoustic sensors: a Doppler Velocity Log (DVL) sensor and an Ultra -Short BaseLine (USBL) device. Afterwards, a solution to autonomously recognise in real-time gas seeps on FLS acoustic imagery utilising a Single Shot MultiBox Detector (SSD) is presented. The relative position of the detected seep with respect to the vehicle is then retrieved and combined with the estimated AUV position to obtain the geodetic location of the seep. Finally, the proposed algorithm was tested during at sea experiments, where gas leaks were artificially reproduced, and the achieved results prove the validity of the proposed method to autonomously detect and accurately geolocate underwater seeps in real time using an AUV.

Automatic target recognition and geolocalisation of natural gas seeps using an autonomous underwater vehicle / Bresciani, Matteo; Zacchini, Leonardo; Topini, Alberto; Ridolfi, Alessandro; Costanzi, Riccardo. - In: CONTROL ENGINEERING PRACTICE. - ISSN 0967-0661. - STAMPA. - 145:(2024), pp. 105864.1-105864.14. [10.1016/j.conengprac.2024.105864]

Automatic target recognition and geolocalisation of natural gas seeps using an autonomous underwater vehicle

Topini, Alberto;Ridolfi, Alessandro;
2024

Abstract

Autonomous Underwater Vehicles (AUVs) have proven to be a precious resource for oceans preservation thanks to their ability to accomplish survey missions on wide areas without the need of human supervision. This also includes the detection and mapping of underwater gas emissions, whether these are due to damaged offshore structures or naturally released from the seafloor. Indeed, submarine seepage have severe repercussions on the surrounding marine habitat and contribute to greenhouse gas emissions, hence the need to identify and monitor them. This work focuses on the online Automatic Target Recognition (ATR) and geolocalisation of underwater gas leakages utilising an AUV equipped with a Forward Looking Sonar (FLS). With the aim to accurately position the seeps, it was necessary to address the problem of navigating the vehicle in areas with the presence of gas bubbles. In fact, bubbles adversely affect acoustic sensors typically exploited by underwater robots for navigation. The paper investigates the effects that seepage have on the performance of navigation solutions based on two different acoustic sensors: a Doppler Velocity Log (DVL) sensor and an Ultra -Short BaseLine (USBL) device. Afterwards, a solution to autonomously recognise in real-time gas seeps on FLS acoustic imagery utilising a Single Shot MultiBox Detector (SSD) is presented. The relative position of the detected seep with respect to the vehicle is then retrieved and combined with the estimated AUV position to obtain the geodetic location of the seep. Finally, the proposed algorithm was tested during at sea experiments, where gas leaks were artificially reproduced, and the achieved results prove the validity of the proposed method to autonomously detect and accurately geolocate underwater seeps in real time using an AUV.
2024
145
1
14
Goal 9: Industry, Innovation, and Infrastructure
Bresciani, Matteo; Zacchini, Leonardo; Topini, Alberto; Ridolfi, Alessandro; Costanzi, Riccardo
File in questo prodotto:
File Dimensione Formato  
1-s2.0-S0967066124000248-main.pdf

accesso aperto

Descrizione: Articolo principale
Tipologia: Pdf editoriale (Version of record)
Licenza: Open Access
Dimensione 3.25 MB
Formato Adobe PDF
3.25 MB Adobe PDF

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1359993
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact