Files in this item



application/pdfGUNGOR-DISSERTATION-2020.pdf (24MB)Restricted Access
(no description provided)PDF


Title:Optimization of lateral position of autonomous truck platoons to improve roadway infrastructure sustainability
Author(s):Gungor, Osman Erman
Director of Research:Al-Qadi, Imad
Doctoral Committee Chair(s):Al-Qadi, Imad
Doctoral Committee Member(s):Ouyang, Yanfeng; Meidani, Hadi; Tutumluer, Erol; Ozer, Hasan; Underwood, Shane
Department / Program:Civil & Environmental Eng
Discipline:Civil Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Autonomous connected and trucks, platooning, sustainability, infrastructure, optimization, pavement
Abstract:The adaptation rate of connected and autonomous vehicle (CAV) technologies correlates with the condition of transportation infrastructure. In fact, an aging and deteriorating infrastructure represents one of the main barriers to advancing CAV technologies. Given that only 41\% of U.S. roads meet the requirements for a ``good ride'', accurately estimating transportation infrastructure performance is critical to ensuring the existence of a functioning and well-maintained transportation infrastructure network for CAV advancement. An accurate representation of traffic input is needed to predict the performance of a transportation infrastructure. Currently, the nature of human drivers characterizes traffic input for transportation analysis and design tools. However, the introduction of CAVs drastically changes the characteristics of such inputs, which in turn may require significant modifications to existing infrastructure design guidelines. Human-driven trucks seldom follow a straight path. Their lateral position within a lane tends to deviate significantly as they travel; this is often referred to as “wheel wander.” While the lateral position of connected and autonomous trucks (CATs) is typically more channelized, it can be remotely controlled. Additionally, the time between two consecutive loadings (or, the “resting period”) is particularly reduced for platooning CATs. Therefore, we need a new pavement design framework to quantify the effects of CATs on pavement structures. This dissertation first addresses this gap by developing a new pavement design framework for CATs. This framework may allow for the modification of analytical pavement design approaches. It does so by explicitly considering loading lateral position and resting periods through a combination of statistical methods and function approximation techniques. After describing this framework, this dissertation presents two control strategies for platooning CATs: de-centralized and centralized. Perfectly aligned trucks at close distances in a platoon would decrease fuel consumption because of reduced aerodynamic drag. However, this may accelerate damage accumulation within pavement structures because of channelized load application and reduced resting time, resulting in a need for more frequent rehabilitation and maintenance activities. Therefore, there is a trade-off between fuel cost savings due to reduced aerodynamic drag and increased pavement life cycle costs. The de-centralized strategy addresses this trade-off by optimizing the lateral positions of the trucks in a single platoon. In other words, it generates an optimum platooning skeleton to improve pavement longevity while preserving fuel efficiency. We assume that such a skeleton is initiated and sustained by vehicle-to-vehicle (V2V) communication as the platoon travels The centralized platooning strategy leverages vehicle-to-infrastructure (V2I) communication to optimize the lateral position of each platoon (or group of platoons on each day or week). In the centralized strategy, the fuel efficiency of trucks is maximized because the trucks in a single platoon are still perfectly aligned. However, applying a centralized strategy may require significant investment because it depends on the existence of reliable, centralized V2I communication, as opposed to the de-centralized strategy’s V2V communication that is already part of the platooning technology. In the last part of this dissertation, we use the developed pavement design framework to demonstrate the importance of incorporating wheel wander while comparing the effects of wide-base tires (WBT) and dual-tire assembly (DTA) on pavement life cycle costs. Historically, research has shown that WBT inflict greater damage on the pavement than do DTAs. However, these two tires have been assumed to have the same wheel wander characteristics. Our results showed that the damage difference induced by these two tires is reduced when wheel wander is incorporated into pavement life cycle cost analysis. In other words, the negative effects of WBT may be overrepresented if wheel wander is not considered.
Issue Date:2020-11-02
Rights Information:2020 Osman Erman Gungor
Date Available in IDEALS:2021-03-05
Date Deposited:2020-12

This item appears in the following Collection(s)

Item Statistics