Date of Award

2014

Degree Type

Thesis

Degree Name

M.A.Sc.

Department

Mechanical, Automotive, and Materials Engineering

First Advisor

Johrendt, Jennifer

Keywords

Artificial Neural Network, Driver Model, Driving Performance, Exploratory Statistics, Risky Behaviour, Unsupervised Classification

Rights

CC BY-NC-ND 4.0

Abstract

Driving performance can be directly related to the driver behaviour in terms of the mental workload and risk perception. No generally accepted model or system exists that can model the driving task or driver performance in a comprehensive manner. The purpose of this research is to develop a methodology using a series of modelling techniques to evaluate driving performance under naturalistic driving contexts. Exploratory statistical techniques and artificial neural network have been used as the backbone of the work presented in this thesis to determine and classify driver performance in different categories by identifying underlying natural sub-sets in the driving data set. A safe and experienced driver should possess the knowledge and the experience about his/her driving skills along with an acute awareness of the surrounding driving environment. The methodology proposed in this thesis can be used for various applications including evaluation of driving performance of emergency ambulance drivers.

Share

COinS