Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Computer Science

Supervisor

Steven Beauchemin

Abstract

Over the last few years, Advanced Driver Assistance Systems (ADAS) have been shown to significantly reduce the number of vehicle accidents. Accord- ing to the National Highway Traffic Safety Administration (NHTSA), driver errors contribute to 94% of road collisions. This research aims to develop a predictive model of driver eye fixation by analyzing the driver eye and head information (cephalo-ocular) for maneuver prediction in an Advanced Driving Assistance System (ADAS). Several ADASs have been developed to help drivers to perform driving tasks in complex environments and many studies were conducted on improving automated systems. Some research has relied on the fact that the driver plays a crucial role in most driving scenarios, recognizing the driver’s role as the central element in ADASs. The way in which a driver monitors the surrounding environment is at least partially descriptive of the driver’s situation awareness. This thesis’s primary goal is the quantitative and qualitative analysis of driver behavior to determine the relationship between driver intent and actions. The RoadLab initiative provided an instrumented vehicle equipped with an on-board diagnostic system, an eye-gaze tracker, and a stereo vision system for the extraction of relevant features from the driver, the vehicle, and the environment. Several driver behavioral features are investigated to determine whether there is a relevant relation between the driver’s eye fixations and the prediction of driving maneuvers.

Summary for Lay Audience

The number of vehicles on our streets and highways increases every day. This fact renders the analysis of traffic situations increasingly complicated. Hence, vehicle manufacturers have been developing Advanced Driver Assistance Systems (ADASs) to avoid 40% of traffic accidents during the driving environment. This research tries to develop a predictive model of driver eye fixation by analyzing the driver eye and head information (cephalo-ocular) for maneuver prediction in an Advanced Driving Assistance System (ADAS). This thesis’s primary goal is the quantitative and qualitative analysis of driver behavior to determine the relationship between driver intent and actions. Several driver behavioral features are investigated to determine whether there is a relevant relationship between the driver’s eye fixations and the prediction of driving maneuvers.

Download

Share

COinS