logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-02042021-163056


Tipo di tesi
Tesi di laurea magistrale
Autore
TIMPERI, ANDREA
URN
etd-02042021-163056
Titolo
Path-planning for AUVs based on Genetic Algorithm for Information Gathering
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
INGEGNERIA ROBOTICA E DELL'AUTOMAZIONE
Relatori
relatore Prof. Caiti, Andrea
relatore Prof. Costanzi, Riccardo
supervisore Prof. Bonin Font, Francisco
Parole chiave
  • Autonomous Underwater Vehicle
  • Gaussian Processes
  • Genetic Algorithm
  • Informative Path Planning
Data inizio appello
25/02/2021
Consultabilità
Non consultabile
Data di rilascio
25/02/2091
Riassunto
In the last decades, sea protection and monitoring have been a central topic in the underwater robotics world. Nowadays, autonomous underwater vehicles (AUVs) usage overcame many mission quality limitations related to inefficient systems employment, but intelligent AUVs use remains the central problem. In this regard, it has been developed a path planner based on a genetic algorithm for information gathering to Posidonia oceanica (PO) meadows monitoring. PO is an endemic plant of the Mediterranean Sea that provides a wide variety of ecosystem services to society, contributing to coastal protection, water quality and fisheries. Unfortunately, due to its coastal location, it suffers numerous human pressures that threaten its conservation, therefore it needs recursive monitoring employing optimised methodologies to increase surveys efficiency and the quality of related information to compare them over time. Hence, the genetic planner moves in a continuous environment without a prefixed destination, establishing the path aim in higher uncertainty areas defined through a Gaussian Process (GP). It will be provided an overview of Informative Path Planning (IPP), GPs and genetic theory, useful for contextualising the planner and genetic operators implemented. Then results, conclusions and future works will allow a suitable framework to appreciate how the work improves information gathering during monitoring surveys.
File