In recent years climate changes effects caused by pollution made interest towards a sustainable mobility ecosystem. In order to enhance efficiency and safety of transport means, advanced systems for driving assistance have been developed, moving a huge step forward in the evolution of the self-driving car. A key aspect for the diffusion of the latter is communication with other vehicles and road users. In order to increase road safety, cars will have to coordinate their own manoeuvres by sharing massive amount of sensor data, with each other vehicle, road infrastructure, and road users through Vehicle To Everything communication. V2X communication represents an unprecedented challenge for telecommunications because of the high mobility and the safety-sensitive environment in which it operates, which calls for a low latency, high reliability and high data rate. One of the most promising solutions to satisfy these requirements is exploiting millimiter wave (mmWave) frequencies which however, due to the frequency in which they operate, are subjected to significant path losses. In order to hinder this drawback effect, narrow communication beams are employed, which concentrate transmission power along the direction pointing towards the receiver. This direction, in a high mobility scenario such as the vehicular one, is subjected to continuous changes and if beams are not steered properly, the communication channel drops. Current beam alignment procedures have been designed for stationary and quasi-stationary conditions and do not satisfy the low latency requirements of V2X communications. In this perspective, this thesis proposes a communication architecture operating on two different layers and exploiting vehicle onboard measurements. Through Global Navigation Satellite System (GNSS), Inertial Measurement Unit (IMU) and other sensors data fusion, each vehicle is able to estimate its own position and orientation. These information is usefully broadcasted in real-time with nearby vehicles to assist the pointing, tracking and steering of the mmWave beam. The performance of this solution is evaluated by simulations over multiple road profiles (according to ISO 8608), evaluating the impact of road pavement irregularities and relative suspension filtering effects, as well as by collecting GPS and IMU data from a dedicated experimental campaign in the city of Milan.
Negli ultimi anni gli effetti dei cambiamenti climatici causati dalle emissioni inquinanti hanno fatto crescere l'interesse verso la mobilità sostenibile. Al fine di migliorare l'efficienza dei mezzi di trasporto e al contempo renderli più sicuri, sono state sviluppati sistemi per la guida assistita, che hanno fatto fare notevoli passi in avanti alla realizzazione dell'auto a guida autonoma. Un aspetto cruciale per la diffusione di quest'ultima è la comunicazione con altri veicoli ed utenti della strada. Per aumentare la sicurezza stradale, i veicoli dovranno coordinare le proprie manovre attraverso uno scambio di dati, con ogni altro, veicolo, infrastruttura stradale ed utente presente sulla strada, tramite la comunicazione Vehicle To Everything (V2X). Quest'ultima rappresenta una sfida senza precedenti per le telecomunicazioni, a causa dell'elevata mobilità e dell'ambiente ad alto rischio in cui opera, che impone vincoli di bassa latenza, elevata affidabilità ed alto data rate. Una delle soluzioni più promettenti per soddisfare tali vincoli è l'utilizzo delle frequenze ad onde millimetriche (mmWave) che tuttavia, proprio a causa della frequenza in cui operano, sono soggette ad elevate perdite lungo il percorso. Per ridurre tale effetto vengono impiegati fasci di comunicazione stretti, che concentrano la potenza di trasmissione verso il dispositivo ricevente. Questa direzione in un ambito ad elevata mobilità come quello veicolare è soggetta a continui cambiamenti, e se i fasci non vengono propriamente indirizzati, il canale di comunicazione cade. I correnti metodi di allineamento dei fasci sono progettati per applicazioni statiche o quasi statiche e non soddisfano i requisiti di bassa latenza richiesti alla comunicazione V2X. A tale scopo, in questa tesi si propone un'architettura di comunicazione a due livelli e che sfrutta le informazioni di sensori a bordo veicolo. Tramite la fusione di misure provenienti dal sistema satellitare globale di navigazione (GNSS), da unità inerziali (IMU) ed altri sensori, ogni veicolo è in grado di stimare la propria posizione ed la propria orientazione. Queste informazioni sono condivise in tempo reale con i veicoli vicini attraverso un canale di comunicazione a bassa frequenza per assistere, di riflesso, il puntamento a onde millimetriche. Le prestazioni di questa metodologia sono validate sia in ambiente simulato per diverse tipologie di strada (in accordo alla normativa ISO 8606), dove viene valutato l'impatto delle irregolarità del manto stradale e il relativo effetto di filtraggio delle sospensioni, sia su dati reali di GPS e IMU acquisiti durante una campagna sperimentale dedicata, svoltasi a Milano.
Sensor-aided V2X beam tracking for connected automated driving
CIARAMITARO, GIOVANNI
2019/2020
Abstract
In recent years climate changes effects caused by pollution made interest towards a sustainable mobility ecosystem. In order to enhance efficiency and safety of transport means, advanced systems for driving assistance have been developed, moving a huge step forward in the evolution of the self-driving car. A key aspect for the diffusion of the latter is communication with other vehicles and road users. In order to increase road safety, cars will have to coordinate their own manoeuvres by sharing massive amount of sensor data, with each other vehicle, road infrastructure, and road users through Vehicle To Everything communication. V2X communication represents an unprecedented challenge for telecommunications because of the high mobility and the safety-sensitive environment in which it operates, which calls for a low latency, high reliability and high data rate. One of the most promising solutions to satisfy these requirements is exploiting millimiter wave (mmWave) frequencies which however, due to the frequency in which they operate, are subjected to significant path losses. In order to hinder this drawback effect, narrow communication beams are employed, which concentrate transmission power along the direction pointing towards the receiver. This direction, in a high mobility scenario such as the vehicular one, is subjected to continuous changes and if beams are not steered properly, the communication channel drops. Current beam alignment procedures have been designed for stationary and quasi-stationary conditions and do not satisfy the low latency requirements of V2X communications. In this perspective, this thesis proposes a communication architecture operating on two different layers and exploiting vehicle onboard measurements. Through Global Navigation Satellite System (GNSS), Inertial Measurement Unit (IMU) and other sensors data fusion, each vehicle is able to estimate its own position and orientation. These information is usefully broadcasted in real-time with nearby vehicles to assist the pointing, tracking and steering of the mmWave beam. The performance of this solution is evaluated by simulations over multiple road profiles (according to ISO 8608), evaluating the impact of road pavement irregularities and relative suspension filtering effects, as well as by collecting GPS and IMU data from a dedicated experimental campaign in the city of Milan.| File | Dimensione | Formato | |
|---|---|---|---|
|
TESI.pdf
non accessibile
Descrizione: 18:51
Dimensione
6.28 MB
Formato
Adobe PDF
|
6.28 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/10589/165474