During the last years, road safety has become a really important topic, and this has led the whole automotive sector to invest in the research and development of ADAS (Advanced Driver Assistance Systems). The situation is in fast evolution: a wide and various offer of these systems already exists on the market and the combination of them is leading to an Autonomous Driving condition. For this reason, this thesis will face with an innovative level 1 Autonomous Driving system developed in previous activities: the Adaptive Speed Limiter (ASL). Its principal and most innovative features are the dependency on the adherence conditions and the ability to provide as output a speed profile, which allows a vehicle to follow a certain trajectory in safety, even in presence of obstacles. First, the ASL logic will be improved, especially from the computational time point of view. Then, considering that up to know it has always been tested through simulations, it will be implemented on a real car through the creation of an interface between the algorithm and the test vehicle. Finally, it will be tested in order to set the thresholds and the parameters needed by the logic to work properly and to point out the benefits for Autonomous Driving systems linked to the knowledge of the potential friction coefficient between tyre and road.
Negli ultimi anni la sicurezza stradale ha assunto un’importanza sempre più rilevante, e ciò ha portato l’intero settore automobilistico ad investire nella ricerca e nello sviluppo di sistemi ADAS (Advanced Driver Assistance Systems). La situazione è in rapida evoluzione: un’ampia e varia offerta di questi sistemi esiste già sul mercato e la loro combinazione sta portando ad una condizione di guida autonoma. Per tale motivo questa tesi si concentra su un sistema innovativo di guida autonoma di livello 1 sviluppato in attività precedenti: l’Adaptive Speed Limiter (ASL). I suoi aspetti principali e maggiormente innovativi sono la dipendenza dalle condizioni di aderenza e la capacità di fornire come output un profilo di velocità che permetta al veicolo di seguire una certa traiettoria in sicurezza, anche in presenza di ostacoli. In primo luogo la logica dell’ASL sarà migliorata, specialmente dal punto di vista del carico computazionale. Poi, considerando che fino ad ora è sempre stato testato attraverso simulazioni, esso verrà implementato su un veicolo reale, grazie alla creazione di un’interfaccia tra l’algoritmo stesso e l’auto di prova. Infine, sarà testato in modo da fissare le soglie e i parametri necessari perché la logica possa funzionare correttamente e per evidenziare i benefici che i sistemi di guida autonoma possono trarre dalla conoscenza del coefficiente d’attrito potenziale tra strada e pneumatico.
Improvement, on board implementation and testing of a new level 1 longitudinal autonomous driving system : the adaptive speed limiter
PIOVESAN, PAOLO
2016/2017
Abstract
During the last years, road safety has become a really important topic, and this has led the whole automotive sector to invest in the research and development of ADAS (Advanced Driver Assistance Systems). The situation is in fast evolution: a wide and various offer of these systems already exists on the market and the combination of them is leading to an Autonomous Driving condition. For this reason, this thesis will face with an innovative level 1 Autonomous Driving system developed in previous activities: the Adaptive Speed Limiter (ASL). Its principal and most innovative features are the dependency on the adherence conditions and the ability to provide as output a speed profile, which allows a vehicle to follow a certain trajectory in safety, even in presence of obstacles. First, the ASL logic will be improved, especially from the computational time point of view. Then, considering that up to know it has always been tested through simulations, it will be implemented on a real car through the creation of an interface between the algorithm and the test vehicle. Finally, it will be tested in order to set the thresholds and the parameters needed by the logic to work properly and to point out the benefits for Autonomous Driving systems linked to the knowledge of the potential friction coefficient between tyre and road.File | Dimensione | Formato | |
---|---|---|---|
Tesi_Paolo_Piovesan.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
6.7 MB
Formato
Adobe PDF
|
6.7 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/10589/138528