Electrical and Computer Engineering ETDs

Publication Date

Summer 7-13-2018

Abstract

Transportation is commonplace around our world. Numerous researchers dedicate great efforts to vast transportation research topics. The purpose of this dissertation is to investigate and address a couple of transportation problems with respect to geographic discretization, pavement surface automatic examination, and traffic ow simulation, using advanced computational technologies. Many applications require a discretized 2D geographic map such that local information can be accessed efficiently. For example, map matching, which aligns a sequence of observed positions to a real-world road network, needs to find all the nearby road segments to the individual positions. To this end, the map is discretized by cells and each cell retains a list of road segments coincident with this cell. An efficient method is proposed to form such lists for the cells without costly overlapping tests. Furthermore, the method can be easily extended to 3D scenarios for fast triangle mesh voxelization. Pavement surface distress conditions are critical inputs for quantifying roadway infrastructure serviceability. Existing computer-aided automatic examination techniques are mainly based on 2D image analysis or 3D georeferenced data set. The disadvantage of information losses or extremely high costs impedes their effectiveness iv and applicability. In this study, a cost-effective Kinect-based approach is proposed for 3D pavement surface reconstruction and cracking recognition. Various cracking measurements such as alligator cracking, traverse cracking, longitudinal cracking, etc., are identified and recognized for their severity examinations based on associated geometrical features. Smart transportation is one of the core components in modern urbanization processes. Under this context, the Connected Autonomous Vehicle (CAV) system presents a promising solution towards the enhanced traffic safety and mobility through state-of-the-art wireless communications and autonomous driving techniques. Due to the different nature between the CAVs and the conventional Human- Driven-Vehicles (HDVs), it is believed that CAV-enabled transportation systems will revolutionize the existing understanding of network-wide traffic operations and re-establish traffic ow theory. This study presents a new continuum dynamics model for the future CAV-enabled traffic system, realized by encapsulating mutually-coupled vehicle interactions using virtual internal and external forces. A Smoothed Particle Hydrodynamics (SPH)-based numerical simulation and an interactive traffic visualization framework are also developed.

Keywords

Intelligent Transportation, Voxelization, Kinect, Connected Autonomous Vehicle, SPH, Traffic Flow Simulation

Document Type

Dissertation

Language

English

Degree Name

Computer Engineering

Level of Degree

Doctoral

Department Name

Electrical and Computer Engineering

First Committee Member (Chair)

Yin Yang

Second Committee Member

Rafael Fierro

Third Committee Member

Marios Pattichis

Fourth Committee Member

Wei Shu

Fifth Committee Member

Guohui Zhang

Download

Share

COinS