Title:
APPLYING COLLABORATIVE ONLINE ACTIVE LEARNING IN VEHICULAR NETWORKS FOR FUTURE CONNECTED AND AUTONOMOUS VEHICLES
APPLYING COLLABORATIVE ONLINE ACTIVE LEARNING IN VEHICULAR NETWORKS FOR FUTURE CONNECTED AND AUTONOMOUS VEHICLES
Author(s)
Liu, Huiye
Advisor(s)
Blough, Douglas M.
Editor(s)
Collections
Supplementary to
Permanent Link
Abstract
The main objective of this thesis is to provide a framework for, and proof of concept of, collaborative online active learning in vehicular networks. Another objective is to advance the state of the art in simulation-based evaluation and validation of connected intelligent vehicle applications. With advancements in machine learning and artificial intelligence, connected autonomous vehicles (CAVs) have begun to migrate from laboratory development and testing conditions to driving on public roads. Their deployment in our environmental landscape offers potential for decreases in road accidents and traffic congestion, as well as improved mobility in overcrowded cities. Although common driving scenarios can be relatively easily solved with classic perception, path planning, and motion control methods, the remaining unsolved scenarios are corner cases in which traditional methods fail. These unsolved cases are the keys to deploying CAVs safely on the road, but they require an enormous amount of data collection and high-quality human annotation, which are very cost-ineffective considering the ever-changing real-world scenarios and highly diverse road/weather conditions. Additionally, evaluating and testing applications for CAVs in real testbeds are extremely expensive, as obvious failures like crashes tend to be rare events and can hardly be captured through predefined test scenarios. Therefore, realistic simulation tools with the benefit of lower cost as well as generating reproducible experiment results are needed to complement the real testbeds in validating applications for CAVs. Therefore, in this thesis, we address the challenges therein and establish the fundamentals of the collaborative online active learning framework in vehicular network for future connected and autonomous vehicles.
Sponsor
Date Issued
2022-05-20
Extent
Resource Type
Text
Resource Subtype
Dissertation