Date of Graduation

2016

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Yaser P Fallah

Committee Co-Chair

Elaine M Eschen

Committee Member

Parviz Famouri

Committee Member

Jason N Gross

Committee Member

Vinod K Kulathumani

Abstract

Recent developments in the area of Connected and Automated Vehicles (CAVs) have boosted the interest in Intelligent Transportation Systems (ITSs). While ITS is intended to resolve and mitigate serious traffic issues such as passenger and pedestrian fatalities, accidents, and traffic congestion; these goals are only achievable by vehicles that are fully aware of their situation and surroundings in real-time. Therefore, connected and automated vehicle systems heavily rely on communication technologies to create a real-time map of their surrounding environment and extend their range of situational awareness. In this dissertation, we propose novel approaches to enhance situational awareness, its applications, and effective sharing of information among vehicles.;The communication technology for CAVs is known as vehicle-to-everything (V2x) communication, in which vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) have been targeted for the first round of deployment based on dedicated short-range communication (DSRC) devices for vehicles and road-side transportation infrastructures. Wireless communication among these entities creates self-organizing networks, known as Vehicular Ad-hoc Networks (VANETs). Due to the mobile, rapidly changing, and intrinsically error-prone nature of VANETs, traditional network architectures are generally unsatisfactory to address VANETs fundamental performance requirements. Therefore, we first investigate imperfections of the vehicular communication channel and propose a new modeling scheme for large-scale and small-scale components of the communication channel in dense vehicular networks. Subsequently, we introduce an innovative method for a joint modeling of the situational awareness and networking components of CAVs in a single framework. Based on these two models, we propose a novel network-aware broadcast protocol for fast broadcasting of information over multiple hops to extend the range of situational awareness. Afterward, motivated by the most common and injury-prone pedestrian crash scenarios, we extend our work by proposing an end-to-end Vehicle-to-Pedestrian (V2P) framework to provide situational awareness and hazard detection for vulnerable road users. Finally, as humans are the most spontaneous and influential entity for transportation systems, we design a learning-based driver behavior model and integrate it into our situational awareness component. Consequently, higher accuracy of situational awareness and overall system performance are achieved by exchange of more useful information.

Download

Share

COinS