Off-campus WSU users: To download campus access dissertations, please use the following link to log into our proxy server with your WSU access ID and password, then click the "Off-campus Download" button below.
Non-WSU users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Access Type
WSU Access
Date of Award
January 2023
Degree Type
Dissertation
Degree Name
Ph.D.
Department
Computer Science
First Advisor
Weisong Shi
Abstract
Since the concept of “edge computing” was proposed in 2015, researchers have devoted themselves to entering a new era, where computation and storage are performed at or near data sources. In the years following, we have witnessed the rapid growth of the connected vehicle (the mobile sensing, computing, communication, and energy storage platform), which is transforming from the vehicle-centric, closed, fixed-function vehicle to the AI-centric, connected, and software-defined vehicle that enable vehicle-to-infrastructure and vehicle-to-vehicle. However, this evolution brings a series of technical challenges (e.g., safety & reliability, vehicular communication latency, and computation constraints).
In this dissertation, we explore a series of research works toward reliable, scalable, and efficient edge-enabled connected vehicle applications. First, we introduce temporal compressive sensing to autonomous driving and propose a vehicle-edge-cloud closed-loop framework with the key advantages of bandwidth reduction and inference acceleration for machine learning models. Then, a collaborative learning framework based on a group of heterogeneous computing platforms will be presented, which enables collaborative training (i.e., electric vehicle battery failure prediction) and collaborative inference (i.e., multi-camera multi-target tracking). Next, to ensure the vehicular data can be stored reliability in the cloud data centers, one of the largest disk failure prediction studies will be introduced. Finally, we summarize this dissertation and discuss promising future work.
Recommended Citation
Lu, Sidi, "Toward Efficient, Scalable, And Reliable Edge-Enabled Applications For Connected Vehicles" (2023). Wayne State University Dissertations. 3860.
https://digitalcommons.wayne.edu/oa_dissertations/3860