Date of Award

2012

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Mechanical, Automotive, and Materials Engineering

Keywords

Driver Model, Driver Risk, Elderly Driver, Neural Network, Objective Method, Vehicle Dynamics

Supervisor

Jennifer L Johrendt

Rights

info:eu-repo/semantics/openAccess

Abstract

As the Baby Boomer generation begins to age along with the advances made in modern medicine, the number of elderly people is expected to increase significantly over the course of the next several decades. As the elderly population increases, the number of elderly drivers on our roads increases as well. Driving is both a physically and cognitively intensive task, and it is a well known fact that as people age, both their physical, and cognitive abilities decrease. As a result, elderly drivers are at an increased risk of being involved in a collision. Currently, the methods to determine driver fitness are limited, and as a result, doctors are placed in a difficult situation where they must choose between protecting their client, the public, and their own reputation; or allowing their client to maintain their accustomed level of independence. While there are elderly drivers who are obviously no longer fit to drive, the problem is making a decision regarding elderly drivers whose ability has not completely deteriorated, and fit in a sort of "gray area". The following research presents the ground work for the development of an objective driver risk assessment tool. The assessment tool makes use of artificial neural networks to both model, and evaluate driver behaviour. Presented herein is the current state of driver modeling, the theory behind neural networking and vehicle dynamics, the process used to develop the model, the performance results, and finally the conclusions that were obtained from the research.

Share

COinS