The following paper was produced by the undergraduate student in collaboration with the Politecnica delle Marche University (UNIVPM) and the company Teoresi Group. Autonomous driving is a crucial research field in the evolution of modern transport systems. The present work aims to investigate the topic of smart mobility, in particular the design of control systems for autonomous driving and the development of systems ADAS, i.e. advanced driver assistance systems, using 3D scenarios for simulation and testing. simulation and testing. In this context, the thesis work focuses on the application of software such as MATLAB®, Simulink®, RoadRunner®, CARLA Simulator, and specific toolboxes for the design of perception, planning and control systems for autonomous driving. The aim is to acquire information on real-world behaviour and verify the functionality of embedded software. In particular, some test cases (use cases) were created in MATLAB-Simulink, focusing on the design of control systems for autonomous vehicles and the integration of ADAS systems and sensors. These test cases are nothing more than simulated scenarios that assess the performance and safety of autonomous driving systems, ensuring that vehicles and sensors can respond correctly to different road situations, perform appropriate manoeuvres or provide adequate measurements. Testing and validation of the test cases in a realistic simulation environment was the crucial phase of the work. For this purpose, the CARLA Simulator was used, which includes predefined realistic maps to test the realised implementation. First, however, the RoadRunner software was employed in order to create a customised map to simulate specific situations and validate some of the implemented test cases. In conclusion, this work contributes to the research and state-of-the-art in the design and testing of autonomous driving systems, exploiting a combination of advanced software tools to design, simulate and test complex scenarios.
Il seguente elaborato è stato realizzato dal laureando in collaborazione con l’Università Politecnica delle Marche (UNIVPM) e l’azienda Teoresi Group. La guida autonoma è un ambito di ricerca cruciale nell’evoluzione dei sistemi di trasporto moderni. Il presente lavoro ha l’intento di approfondire il tema della smart mobility, in particolare la progettazione di sistemi di controllo per la guida autonoma e lo sviluppo di sistemi ADAS, cioè sistemi avanzati di assistenza al conducente, avvalendosi di scenari 3D per la simulazione e il testing. In questo contesto, il lavoro di tesi si focalizza sull'applicazione di software quali MATLAB®, Simulink®, RoadRunner®, CARLA Simulator, e toolbox specifici per la progettazione di sistemi di percezione, pianificazione e controllo per la guida autonoma. L'obiettivo è quello di acquisire informazioni sul comportamento del mondo reale e verificare la funzionalità del software embedded. In particolare, sono stati creati alcuni test cases (casi d’uso) in MATLAB-Simulink, focalizzati sulla progettazione di sistemi di controllo per veicoli autonomi e sull'integrazione di sistemi ADAS e sensori. Tali test cases non sono altro che scenari simulati che valutano le prestazioni e la sicurezza dei sistemi di guida autonoma, assicurando che i veicoli e i sensori possano rispondere correttamente a diverse situazioni stradali, compiere manovre adeguate o fornire misurazioni adeguate. Il testing e la validazione dei test cases in un ambiente di simulazione realistico sono stati la fase cruciale del lavoro. Per questa finalità, è stato impiegato CARLA Simulator, il quale include mappe realistiche predefinite per testare l'implementazione realizzata. Innanzitutto, è stato però impiegato il software RoadRunner al fine di creare una mappa personalizzata per simulare specifiche situazioni e validare alcuni dei test cases implementati. In conclusione, questo lavoro contribuisce alla ricerca e allo stato dell’arte nella progettazione e verifica dei sistemi di guida autonoma, sfruttando una combinazione di strumenti software avanzati per progettare, simulare e testare scenari complessi.
Progettazione di sistemi di controllo per la guida autonoma e definizione e creazione di scenari 3D per il testing
CATALINI, FEDERICO
2022/2023
Abstract
The following paper was produced by the undergraduate student in collaboration with the Politecnica delle Marche University (UNIVPM) and the company Teoresi Group. Autonomous driving is a crucial research field in the evolution of modern transport systems. The present work aims to investigate the topic of smart mobility, in particular the design of control systems for autonomous driving and the development of systems ADAS, i.e. advanced driver assistance systems, using 3D scenarios for simulation and testing. simulation and testing. In this context, the thesis work focuses on the application of software such as MATLAB®, Simulink®, RoadRunner®, CARLA Simulator, and specific toolboxes for the design of perception, planning and control systems for autonomous driving. The aim is to acquire information on real-world behaviour and verify the functionality of embedded software. In particular, some test cases (use cases) were created in MATLAB-Simulink, focusing on the design of control systems for autonomous vehicles and the integration of ADAS systems and sensors. These test cases are nothing more than simulated scenarios that assess the performance and safety of autonomous driving systems, ensuring that vehicles and sensors can respond correctly to different road situations, perform appropriate manoeuvres or provide adequate measurements. Testing and validation of the test cases in a realistic simulation environment was the crucial phase of the work. For this purpose, the CARLA Simulator was used, which includes predefined realistic maps to test the realised implementation. First, however, the RoadRunner software was employed in order to create a customised map to simulate specific situations and validate some of the implemented test cases. In conclusion, this work contributes to the research and state-of-the-art in the design and testing of autonomous driving systems, exploiting a combination of advanced software tools to design, simulate and test complex scenarios.File | Dimensione | Formato | |
---|---|---|---|
TESI_LAUREA_MAGISTRALE_CATALINI_FEDERICO_pdfA.pdf
embargo fino al 19/08/2025
Descrizione: File PDF/A documento tesi Federico Catalini
Dimensione
5.94 MB
Formato
Adobe PDF
|
5.94 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.12075/16571