Hydroplaning poses a high risk in road safety and is the main cause of accidents in bad weather conditions, since it leads to a loss of traction that prevents the vehicle from responding to the driver inputs. The risks involved in hydroplaning conditions led to an increased interest by the automotive industry in the development of innovative solutions to detect this phenomenon. This thesis presents two hydroplaning detection algorithms developed on a rear wheel drive vehicle that, exploiting the measures of stock vehicle sensors and smart tires, aim at estimating an hydroplaning level for each front tire. Both algorithms relies on effect-based detection methods, which are designed to detect the effects of hydroplaning on the vehicle taking into account the longitudinal and the lateral dynamics. The first hydroplaning detection algorithm proposed in this dissertation combines various effect-based methods to estimate hydroplaning using only vehicle signals measured by stock sensors. The second is an evolution of the first: it integrates the effect-based detection methods with the hydroplaning estimation of smart tires, which must be mounted on the vehicle, exploiting the potential of both approaches. Its goal is to overcome the smart tires reliability issues and achieve an accurate and early detection of hydroplaning. Both algorithms have been extensively tested in an experimental campaign and an in-depth analysis on the detection results is presented.
Il fenomeno dell'aquaplaning rappresenta un rischio elevato per la sicurezza stradale ed è la principale causa di incidenti in caso di maltempo, poichè provoca una perdita di trazione che impedisce al veicolo di rispondere ai comandi del pilota. I rischi legati al fenomeno dell'aquaplaning hanno portato ad un crescente interesse da parte dell'industria automobilistica allo sviluppo di soluzioni innovative per identificare questo fenomeno. Questa tesi propone due algoritmi di stima di aquaplaning sviluppati per veicoli a trazione posteriore che, sfruttando segnali misurati da sensori di serie del veicolo e dalle smart tires, mirano a stimare un livello di aquaplaning per ciascuna ruota anteriore. Entrambi gli algoritmi si basano su metodi effect-based, sviluppati per identificare gli effetti dell'aquaplaning sul veicolo tenendo conto sia della dinamica longitudinale che di quella laterale. Il primo algoritmo di stima presentato in questa tesi combina alcuni metodi effect-based per stimare l'aquaplaning, sfruttando solo segnali veicolo misurati con sensori di serie. Il secondo, invece, è un'evoluzione del primo e si basa sull'integrazione dei metodi effect-based con la stima di aquaplaning delle smart tires, sfruttando le potenzialità di entrambi gli approcci. L'obbiettivo di quest'ultimo è il superamento dei problemi di affidabilità delle smart tires per ottenere una stima di aquaplaning accurata e tempestiva. Entrambi gli algoritmi sono stati testati ampiamente in una campagna sperimentale, i cui risultati sono analizzati approfonditamente in questa trattazione.
Analysis and development of effect-based hydroplaning detection algorithms
Canzi, Davide
2021/2022
Abstract
Hydroplaning poses a high risk in road safety and is the main cause of accidents in bad weather conditions, since it leads to a loss of traction that prevents the vehicle from responding to the driver inputs. The risks involved in hydroplaning conditions led to an increased interest by the automotive industry in the development of innovative solutions to detect this phenomenon. This thesis presents two hydroplaning detection algorithms developed on a rear wheel drive vehicle that, exploiting the measures of stock vehicle sensors and smart tires, aim at estimating an hydroplaning level for each front tire. Both algorithms relies on effect-based detection methods, which are designed to detect the effects of hydroplaning on the vehicle taking into account the longitudinal and the lateral dynamics. The first hydroplaning detection algorithm proposed in this dissertation combines various effect-based methods to estimate hydroplaning using only vehicle signals measured by stock sensors. The second is an evolution of the first: it integrates the effect-based detection methods with the hydroplaning estimation of smart tires, which must be mounted on the vehicle, exploiting the potential of both approaches. Its goal is to overcome the smart tires reliability issues and achieve an accurate and early detection of hydroplaning. Both algorithms have been extensively tested in an experimental campaign and an in-depth analysis on the detection results is presented.File | Dimensione | Formato | |
---|---|---|---|
Canzi_Thesis.pdf
non accessibile
Descrizione: Thesis
Dimensione
50.53 MB
Formato
Adobe PDF
|
50.53 MB | Adobe PDF | Visualizza/Apri |
Canzi_ExecutiveSummary.pdf
non accessibile
Descrizione: Executive Summary
Dimensione
2.22 MB
Formato
Adobe PDF
|
2.22 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/10589/201493