The objective of this thesis is the development and experimentation of advanced path planning algorithms, inspired by established techniques such as A* and Rapidly-exploring Random Tree (RRT), for applications on autonomous marine vehicles. The work focuses particularly on the implementation for BlueRobotics’ Blue Boat, and more generally, it is adaptable to marine vehicles used for both research and commercial purposes. Specifically, the development focuses on generating paths and obstacle avoidance approaches for static and known obstacles. Analyses and considerations regarding the comparison of results obtained from tests in both simulated and real environments for both algorithms are presented. A critical aspect of the project is the idea of integrating COLREG (Collision Regulations), which govern ship routes to prevent collisions at sea. The developed system aims to plan optimal paths and adapt them in real-time to ensure compliance with these safety regulations. The validation of the algorithms and their integration was carried out through detailed simulations and practical tests in a controlled environment, highlighting the system's effectiveness in navigating safely and efficiently. The results obtained show a system capable of managing autonomous navigation and paving the way for possible developments in the field of autonomous marine vehicles.
L’obiettivo di questa tesi è lo sviluppo e la sperimentazione di algoritmi di path planning avanzati, ispirati a tecniche consolidate come A* e Rapidly-exploring Random Tree (RRT), per applicazioni su veicoli marini autonomi. Il lavoro si focalizza in particolare sull’implementazione per la Blue Boat di BlueRobotics e, in generale, adattabile a veicoli marini utilizzati sia per la ricerca che per scopi commerciali. In particolare lo sviluppo si concentra sulla generazione di percorsi e approcci di aggiramento di ostacoli statici e conosciuti a priori. Vengono riportate analisi e considerazioni in merito alla comparazione dei risultati ottenuti da test con ambiente di simulazione e in ambiente reale di entrambi gli algoritmi. Un aspetto critico del progetto è l’idea di integrazione delle normative COLREG (Collision Regulations), che regolano la rotta delle navi per prevenire collisioni in mare. Il sistema sviluppato mira a pianificare percorsi ottimali e adattarli in tempo reale per garantire la conformità a queste norme di sicurezza. La validazione degli algoritmi e dell’integrazione è stata effettuata tramite simulazioni dettagliate e test pratici in ambiente controllato, evidenziando l’efficacia del sistema nel navigare in maniera sicura ed efficiente. I risultati ottenuti mostrano un sistema in grado di gestire la navigazione autonoma e in grado di aprire la strada a possibili sviluppi nel campo dei veicoli marini autonomi.
Implementazione e test di algoritmi di path planning per veicoli marini nel rispetto delle norme COLREG
GALASSI, MATTIA
2023/2024
Abstract
The objective of this thesis is the development and experimentation of advanced path planning algorithms, inspired by established techniques such as A* and Rapidly-exploring Random Tree (RRT), for applications on autonomous marine vehicles. The work focuses particularly on the implementation for BlueRobotics’ Blue Boat, and more generally, it is adaptable to marine vehicles used for both research and commercial purposes. Specifically, the development focuses on generating paths and obstacle avoidance approaches for static and known obstacles. Analyses and considerations regarding the comparison of results obtained from tests in both simulated and real environments for both algorithms are presented. A critical aspect of the project is the idea of integrating COLREG (Collision Regulations), which govern ship routes to prevent collisions at sea. The developed system aims to plan optimal paths and adapt them in real-time to ensure compliance with these safety regulations. The validation of the algorithms and their integration was carried out through detailed simulations and practical tests in a controlled environment, highlighting the system's effectiveness in navigating safely and efficiently. The results obtained show a system capable of managing autonomous navigation and paving the way for possible developments in the field of autonomous marine vehicles.| File | Dimensione | Formato | |
|---|---|---|---|
|
Galassi_Mattia.pdf
accesso aperto
Dimensione
18.58 MB
Formato
Adobe PDF
|
18.58 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/71294