Situation-aware Customized Machine Intelligence for Transportation Safety, Equity, and Resilience

relationships.isAuthorOf

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Urbanization brings significant opportunities for improved quality of life, but it also poses complex challenges in transportation, such as safety, efficiency, equity, and privacy concerns. The widespread deployment of smart city sensors, along with data from mobile devices like onboard sensors and smartphones, has created a substantial "big data" environment. This dissertation harnesses this vast amount of data to develop a connected and autonomous transportation system that enhances safety, equity, and resilience through the adaptation and customization of machine intelligence. More specifically, the dissertation introduces a three-tiered strategy to customize machine intelligence in transportation scenarios. Firstly, at the data collection stage, it employs cyber-physical collaboration to integrate road environment and situational data into the sensing framework to enhance situational awareness, thereby improving data accuracy and trustworthiness. Then, for data processing and modeling, integrated sensing technologies are harnessed, synthesizing inputs from various sensors to provide a detailed and comprehensive understanding of the traffic scene. Finally, at the application level, the dissertation presents a human-machine interaction framework, utilizing advanced communication technologies to design customized traffic warning, control, and management systems responsive to diverse user needs across various scenarios. Additionally, the research enhances the efficiency, utility, reliability, and privacy of machine intelligence by integrating customized systems with cutting-edge distributed computing, extending benefits to a wide array of settings, including underserved rural and low-income areas. In summary, the dissertation introduces an innovative, situation-aware machine intelligence system that utilizes distributed computing technology to deliver real-time, reliable, and personalized traffic services. This system upholds safety, equity, and resilience, ensuring equitable service across the transportation field.

Description

Thesis (Ph.D.)--University of Washington, 2024

Citation

DOI