ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-04222019-145426


Tipo di tesi
Tesi di dottorato di ricerca
Autore
MANZARI, VINCENZO
URN
etd-04222019-145426
Titolo
INCREASING THE AUTONOMY OF UNDERWATER VEHICLES OPERATIONS: LONG-ENDURANCE AND PRECISE NAVIGATION PERSPECTIVES
Settore scientifico disciplinare
ING-INF/04
Corso di studi
INGEGNERIA DELL'INFORMAZIONE
Relatori
tutor Prof. Caiti, Andrea
tutor Prof. Luise, Marco
tutor C.F. (AN) Stifani, Mirko
controrelatore Prof. Allotta, Benedetto
controrelatore Prof. Bonin Font, Francisco
Parole chiave
  • passive tracking
  • marine robotics
  • wave energy harvesting
  • Autonomy
  • Autonomous Underwater Vehicle
  • bearing-only
Data inizio appello
09/05/2019
Consultabilità
Completa
Riassunto
AUV (Autonomous Underwater Vehicle) technology began its development in the 1970s.
Since then, improvements in energy efficiency, sensors technology, and computational power have fostered their wide application. As a result, many complex missions that were originally accomplished with towed sensors or manned vehicles are being automated at different levels. AUV designs include torpedo-shapes, gliders, and hovering capable vehicles, and their sizes range from human-portable to hundreds of tons. AUVs are now used for a variety of tasks, including oceanographic surveys, mine countermeasure, anti-submarine warfare, and bathymetric data collection in marine environments from rivers to open oceans. In particular, vehicle endurance is important and crucial because it affects the working duration and mission completion. Research in augmenting AUV persistence has burst in the last ten years, and several techniques address this issue. Among them, environmental energy harvesting is the most promising method. Moreover, AUV accurate localization and navigation are essential to reliably determine its position and to ensure the correct georeferentiation of the senses data for critical applications.

The thesis develops these two perspectives of AUVs operations: long-endurance and precise navigation.
First, a unique prototype, called WAVE (Wave-powered Autonomous Vehicle for marine Exploration), for both energy harvesting from the wave motion and low consumption propulsion, is experimentally characterized. Second, an easily deployable Underwater Test Range (UTR) for Verification and Validation (V&V) of AUVs autonomous navigation skills is presented. These developed capabilities are validated through extensive field demonstrations thanks to SEALab joint applied research laboratory on heterogeneous and autonomous marine systems. It was established in 2015 as a result of an agreement between the Naval Experimentation and Support Center (CSSN) of the Italian Navy and the Interuniversity Research Center of Integrated Systems for Marine Environment (ISME), of which the University of Pisa is a co-founder.

The WAVE module, installed on the modular, torpedo-shaped, hybrid glider/AUV Fòlaga, undergoes an extensive experimental campaign in a towing tank equipped with a wave generator. During the three days of experimentation, a considerable quantity of data related to different recreated sea conditions and WAVE module configurations was collected. From the energetic characterization of the proposed system in terms of average generated power, the most effective configuration of the WAVE module for the battery charging was identified. Preliminary data processing allowed a first tuning of the designed system dynamical model and the simulative results of the expected performance of WAVE under typical Mediterranean sea conditions not tested in the towing tank are presented in this thesis. More tests with various WAVE module configurations are ongoing to enable deeper analysis and better fitting of the modelling parameters in order to be able to test innovative wing profiles in more complex wave scenarios.

A first experimental assessment of the UTR demonstrates the conceived system based on bearing-only sensors, which is capable of precisely tracking an Autonomous Surface Vehicle (ASV) equipped with Differential GPS as position ground truth. With roots in previous simulative results, estimation of target trajectory is performed via non-linear Kalman Filtering (KF) approaches: both Extended and Unscented KF versions have been implemented and compared. Results show how the proposed methodology performs in a real marine scenario with challenging conditions due to shallow waters and magnetically noisy environment. Future steps will aim to enhance the tracking performance using moving sensor platforms to optimize the UTR topology and implementing a distributed approach with mixed bearing-range measurements.

The results reported in the thesis show how AUV can be useful in not traditional and complex scenarios. In the imminent future, AUVs will be increasingly used in long-term, long-range, deep-water missions, operating without the support of a ship-based acoustic positioning system. This will require both precise navigation systems, independent of external aids, and an effective energy system that properly combines the primary batteries of the vehicle with environmental energy sources.
File