Tesi etd-03212025-091612 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
LORENZI, GABRIELE
URN
etd-03212025-091612
Titolo
Generazione dinamica di percorsi 2D per un AGV sulla base di primitive geometriche di tipo η3-spline
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
INGEGNERIA ROBOTICA E DELL'AUTOMAZIONE
Relatori
relatore Prof.ssa Pallottino, Lucia
tutor Ing. Ghilardelli, Fabio
tutor Ing. Ghilardelli, Fabio
Parole chiave
- curve fitting
- eta3-splines
- polynomial splines
- smooth path generation
Data inizio appello
09/04/2025
Consultabilità
Completa
Riassunto
Il contesto industriale moderno, caratterizzato da ambienti dinamici, ha spinto le aziende a sviluppare soluzioni innovative per i veicoli automatici (AGV), che devono essere in grado di adattarsi rapidamente a situazioni impreviste, come ostacoli mobili o la necessità di modificare i percorsi in tempo reale. E80 Group S.P.A., azienda leader nella progettazione di sistemi per la gestione automatizzata dei flussi di materiali, affronta queste sfide sviluppando soluzioni avanzate per migliorare la reattività e la flessibilità degli AGV in ambienti complessi. La tesi si concentra sulla creazione di algoritmi capaci di generare percorsi ottimali, ma anche adattabili alle circostanze, risolvendo i problemi legati alla discontinuità dei percorsi tradizionali basati su waypoint discreti. Gli approcci sviluppati garantiscono un movimento fluido e sicuro, riducendo l'usura degli attuatori e migliorando l'efficienza operativa. La metodologia proposta si basa su curve parametriche, in particolare le η3-spline, che assicurano una continuità geometrica di ordine 3. Sono stati sviluppati due approcci principali con varie tecniche: uno basato sullo "smoothing" dei percorsi esistenti, per generare dei percorsi continui, e l’altro sul fitting di curve continue che si adattano ai waypoint. Entrambi gli approcci sono stati implementati in un plug-in per il sistema ROS2 e validati tramite simulazioni e test su un robot fisico, il Segway RMP Lite 220, con risultati positivi in termini di qualità e affidabilità. La tesi si conclude con un’analisi dei vantaggi e delle possibili evoluzioni future delle tecniche proposte, evidenziando il potenziale per ottimizzare ulteriormente la gestione dei percorsi in ambienti complessi.
The modern industrial context, characterized by dynamic environments, has pushed companies to develop innovative solutions for automated guided vehicles (AGVs), which must be able to quickly adapt to unexpected situations, such as moving obstacles or the need to modify paths in real time. E80 Group S.P.A., a leading company in the design of systems for automated material flow management, faces these challenges by developing advanced solutions to improve the responsiveness and flexibility of AGVs in complex environments. This thesis focuses on creating algorithms capable of generating optimal and adaptable paths, solving issues related to the discontinuity of traditional paths based on discrete waypoints. The developed approaches ensure smooth and safe movement, reducing actuator wear and improving operational efficiency. The proposed methodology is based on parametric curves, particularly η3-splines, which ensure third-order geometric continuity. Two main approaches with various techniques were developed: the first is based on smoothing existing paths to generate continuous paths; the second is based on fitting continuous curves that adapt to waypoints. Both approaches were implemented in a plug-in for the ROS2 system and validated through simulations and tests on a physical robot, the Segway RMP Lite 220, with positive results in terms of quality and reliability. The thesis concludes with an analysis of the advantages and potential future developments of the proposed techniques, highlighting the potential to further optimize path management in complex environments.
The modern industrial context, characterized by dynamic environments, has pushed companies to develop innovative solutions for automated guided vehicles (AGVs), which must be able to quickly adapt to unexpected situations, such as moving obstacles or the need to modify paths in real time. E80 Group S.P.A., a leading company in the design of systems for automated material flow management, faces these challenges by developing advanced solutions to improve the responsiveness and flexibility of AGVs in complex environments. This thesis focuses on creating algorithms capable of generating optimal and adaptable paths, solving issues related to the discontinuity of traditional paths based on discrete waypoints. The developed approaches ensure smooth and safe movement, reducing actuator wear and improving operational efficiency. The proposed methodology is based on parametric curves, particularly η3-splines, which ensure third-order geometric continuity. Two main approaches with various techniques were developed: the first is based on smoothing existing paths to generate continuous paths; the second is based on fitting continuous curves that adapt to waypoints. Both approaches were implemented in a plug-in for the ROS2 system and validated through simulations and tests on a physical robot, the Segway RMP Lite 220, with positive results in terms of quality and reliability. The thesis concludes with an analysis of the advantages and potential future developments of the proposed techniques, highlighting the potential to further optimize path management in complex environments.
File
Nome file | Dimensione |
---|---|
tesi_Lorenzi.pdf | 12.00 Mb |
Contatta l’autore |