ºìÐÓÊÓƵ

Topic

Intelligent Microscopic Models for Traffic Flow Characterization

Department of Electrical and Computer Engineering

Date & location

• Tuesday, April 22, 2025

• 11:00 A.M.

• Virtual Defence

Reviewers

Supervisory Committee

• Dr. T. Aaron Gulliver, Department of Electrical and Computer Engineering, University of Victoria (Co-Supervisor)

• Dr. Zawar Khan, Department of Electrical and Computer Engineering, UVic (Co-Supervisor)

• Dr. Phalguni Mukhopadhyaya, Department of Civil Engineering, UVic (Outside Member) 

External Examiner

• Dr. Eric J. Miller, Department of Civil & Mineral Engineering, University of Toronto 

Chair of Oral Examination

• Dr. Laura Minet, Department of Civil Engineering, UVic

   

Abstract

In this dissertation, microscopic models for traffic flow characterization are studied. Based on the traffic flow evolution characteristics and aiming to characterize the traffic behavior accurately and realistically, this research focuses on developing realistic traffic flow models and enhancing traffic safety, efficiency and pollution control. A traffic model based on driver response is introduced considering both driver reaction and sensitivity. Driver sensitivity includes typical, sluggish, or aggressive drivers. Then the pavement condition is investigated using the pavement condition index (PCI). The impact of fog on visibility is a major factor affecting traffic congestion and safety. Thus, the traffic behavior based on visibility during foggy weather is also investigated. In addition, the recent introduction of connected autonomous vehicles (CAVs) has had a significant impact on road networks. Therefore, a spring-mass based traffic model to evaluate human-driven vehicles (HVs), autonomous vehicles (AVs), and CAVs behavior on a horizontal curve is proposed. Further, the CAVs behavior at bottlenecks considering the cyberattack is investigated. This dissertation also develops an energy consumption model considering driver energy saving awareness. The performance of traffic models is presented and compared with the intelligent driver (ID) model, and traffic stability is analyzed. The results demonstrate the advantages of our approach.