Multirotor Unmanned Aerial Vehicles (UAVs) have been gaining increasing interest and importance in the last few years due to their flexibility and effectiveness across a wide range of applications. In particular formations of multi-drones is a topic widely studied in the literature. Formation control refers to the design of control strategies that enable a group of agents to maintain a desired geometric configuration while tracking a desired trajectory or performing coordinated tasks. This thesis focuses on the problem of fleet formation control. Two strategies are considered: displacement-based control and distance-based control. Firstly, a comprehensive analysis of graph theoretic concepts and rigidity theory is provided to establish the theoretical foundation. The two control laws are then analyzed separately. In distance-based formation control, the target geometry is described solely through inter-agent distances. However, distances alone do not convey any information about the global orientation of the formation: two configurations may have exactly the same pairwise distances while being mirror images of each other. As a consequence, the agents may converge either to the intended shape or to its reflected counterpart. This phenomenon is referred to as flip ambiguity. To overcome this issue, a bearing control term is added. Moreover, the position control term is modified to enhance stability in case of scaling of the formation. To ensure safety during motion, Control Barrier Functions (CBF) and Exponential Control Barrier Functions (ECBF) are included as a safety filter to avoid collisions between drones and maintain them in a prescribed safe area. The proposed methodologies are tested through both simulations and real-world experiments.
Gli Aeromobili a Pilotaggio Remoto (APR) multirotore hanno acquisito sempre più interesse e importanza negli ultimi anni grazie alla loro flessibilità ed efficacia in un’ampia gamma di applicazioni. In particolare, la gestione di formazioni di droni rappresenta un argomento ampiamente studiato in letteratura. Il controllo di formazione si riferisce alla progettazione di strategie di controllo che consentano a un gruppo di agenti di mantenere una configurazione geometrica desiderata mentre seguono una traiettoria o svolgono compiti coordinati. Questa tesi si concentra sul problema del controllo di formazione per f lotte di droni, considerando due strategie principali: il controllo basato sugli spostamenti e quello basato sulle distanze. Inizialmente viene fornita un’analisi teorica approfondita dei concetti di teoria dei grafi e teoria della rigidità, al fine di stabilire le basi teoriche. Le due leggi di controllo vengono quindi analizzate separatamente. Nel controllo basato sulla distanza la formazione è ottenuta attraverso le distanze tra agenti. Queste, però, non contengono informazioni riguardanti l’orientazione della formazione: due configurazioni possono avere le stesse distanze, ma essere una la riflessione dell’altra. Di conseguenza gli agenti posso convergere o alla forma desiderata o alla sua riflessa. Questo fenomeno è chiamato ambiguità di riflessione. Per risolvere questo problema è stato aggiunto un controllo di orientazione. Inoltre il controllo di posizione è modificato per migliorare la stabilità nel caso in cui la formazione sia scalata. Per garantire la sicurezza durante la missione, vengono integrate nel modello le Control Barrier Functions (CBF) e Exponential Control Barrier Functions (ECBF), utilizzate come filtri di sicurezza per evitare collisioni tra i droni e mantenerli all’interno di un’area sicura prestabilita. Le metodologie proposte sono testate tramite simulazioni ed esperimenti reali.
Quadrotor formation flight strategies with collision avoidance capabilities
Bonacina, Matteo
2024/2025
Abstract
Multirotor Unmanned Aerial Vehicles (UAVs) have been gaining increasing interest and importance in the last few years due to their flexibility and effectiveness across a wide range of applications. In particular formations of multi-drones is a topic widely studied in the literature. Formation control refers to the design of control strategies that enable a group of agents to maintain a desired geometric configuration while tracking a desired trajectory or performing coordinated tasks. This thesis focuses on the problem of fleet formation control. Two strategies are considered: displacement-based control and distance-based control. Firstly, a comprehensive analysis of graph theoretic concepts and rigidity theory is provided to establish the theoretical foundation. The two control laws are then analyzed separately. In distance-based formation control, the target geometry is described solely through inter-agent distances. However, distances alone do not convey any information about the global orientation of the formation: two configurations may have exactly the same pairwise distances while being mirror images of each other. As a consequence, the agents may converge either to the intended shape or to its reflected counterpart. This phenomenon is referred to as flip ambiguity. To overcome this issue, a bearing control term is added. Moreover, the position control term is modified to enhance stability in case of scaling of the formation. To ensure safety during motion, Control Barrier Functions (CBF) and Exponential Control Barrier Functions (ECBF) are included as a safety filter to avoid collisions between drones and maintain them in a prescribed safe area. The proposed methodologies are tested through both simulations and real-world experiments.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_12_Bonacina_Tesi.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: testo tesi
Dimensione
8.34 MB
Formato
Adobe PDF
|
8.34 MB | Adobe PDF | Visualizza/Apri |
|
2025_12_Bonacina_Executive Summary.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Executive Summary
Dimensione
1.9 MB
Formato
Adobe PDF
|
1.9 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/246603