In MotoGP, having a detailed scan of circuits is a significant advantage for the preparation, analysis and optimisation of performance. Currently, the acquisition of this data is done manually, but Ducati aims to automate the process, while increasing the accuracy of the information collected. This thesis, carried out in collaboration with Ducati Motor Holding S.p.A., in the Ducati Corse department, aims to develop a virtual simulation environment for an autonomous differential robot designed for 3D scanning of MotoGP circuits. The main objective is to ensure high accuracy in trajectory tracking by the robot in order to obtain an accurate scan of the track. To test and optimise the robot's navigation, the Gazebo platform was used, integrated with the Matlab-Simulink environment. In particular, a realistic model of the robot and the environment was created from CAD drawings. Subsequently, an algorithm was developed in Simulink to guide the robot along a trajectory defined by GPS points, known a priori. The simulations were conducted in a virtual environment on scenarios of varying complexity, with particular attention to the management of position errors, ensuring that the robot can in future operate autonomously and accurately in real-world contexts. The results obtained confirm the effectiveness of the proposed approach.
In ambito MotoGP, disporre di una scansione dettagliata dei circuiti rappresenta un vantaggio significativo per la preparazione, l’analisi e l’ottimizzazione delle performance. Attualmente, l’acquisizione di questi dati avviene manualmente, ma Ducati punta ad automatizzare il processo, incrementando al contempo la precisione delle informazioni raccolte. Questa tesi, svolta in collaborazione con Ducati Motor Holding S.p.A., nel reparto Ducati Corse, si propone di sviluppare un ambiente di simulazione virtuale per un robot autonomo differenziale progettato per la scansione 3D dei circuiti MotoGP. L’obiettivo principale è garantire un’elevata precisione nell’inseguimento di traiettorie da parte del robot, al fine di ottenere una scansione accurata del tracciato. Per testare e ottimizzare la navigazione del robot, è stata utilizzata la piattaforma Gazebo, integrata con l’ambiente Matlab-Simulink. In particolare, è stato creato un modello realistico del robot e dell’ambiente a partire da disegni CAD. Successivamente, è stato sviluppato in Simulink un algoritmo in grado di guidare il robot lungo una traiettoria definita da punti GPS, noti a priori. Le simulazioni sono state condotte in un ambiente virtuale su scenari di diversa complessità, con particolare attenzione alla gestione degli errori di posizione, garantendo che il robot possa in futuro operare in modo autonomo e preciso anche in contesti reali. I risultati ottenuti confermano l’efficacia dell’approccio proposto.
Simulazione di un robot autonomo per la scansione 3D dei circuiti MotoGP
MARAN, SOFIA
2024/2025
Abstract
In MotoGP, having a detailed scan of circuits is a significant advantage for the preparation, analysis and optimisation of performance. Currently, the acquisition of this data is done manually, but Ducati aims to automate the process, while increasing the accuracy of the information collected. This thesis, carried out in collaboration with Ducati Motor Holding S.p.A., in the Ducati Corse department, aims to develop a virtual simulation environment for an autonomous differential robot designed for 3D scanning of MotoGP circuits. The main objective is to ensure high accuracy in trajectory tracking by the robot in order to obtain an accurate scan of the track. To test and optimise the robot's navigation, the Gazebo platform was used, integrated with the Matlab-Simulink environment. In particular, a realistic model of the robot and the environment was created from CAD drawings. Subsequently, an algorithm was developed in Simulink to guide the robot along a trajectory defined by GPS points, known a priori. The simulations were conducted in a virtual environment on scenarios of varying complexity, with particular attention to the management of position errors, ensuring that the robot can in future operate autonomously and accurately in real-world contexts. The results obtained confirm the effectiveness of the proposed approach.File | Dimensione | Formato | |
---|---|---|---|
Maran_Sofia.pdf
accesso riservato
Dimensione
7.56 MB
Formato
Adobe PDF
|
7.56 MB | Adobe PDF |
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/20.500.12608/84958