Interactive Autonomy in Mixed Traffic: Modeling, Control and Optimization

Date

2023-08

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Enabled by advanced sensing, actuating, and communication technology, connected and automated vehicles (CAVs) have significant potential to improve the efficiency and sustainability of today’s urban transportation system. Despite the growing vehicle connectivity and autonomy in recent years, human drivers will remain to be the majority of traffic participants, and mixed traffic where human-driven vehicles and CAVs coexist is expected to be the most likely traffic scenario in the foreseeable future. There are still some challenges that need to be addressed before CAVs become widely adopted, including public acceptance, technological limitations, and interoperability between CAVs and human-driven vehicles. Therefore, it is essential to develop human-centric CAV systems by placing humans at the center of the CAV system development in mixed traffic scenarios. This dissertation aims to develop modeling, control, and optimization methods for human-centric CAVs to operate in interactive mixed traffic scenarios for improving the efficiency of the hybrid ground transportation systems while acknowledging the surrounding human drivers’ unique driving behaviors and social interactions. First, driver behavior models are designed to understand the drivers’ unique and rich driving preferences in interactive mixed traffic scenarios. Second, the personalized, energy-efficient, and traffic-friendly control strategies for the CAVs featuring the driver behavior models and vehicle connectivity are implemented to improve the efficiency and sustainability of the mixed traffic scenarios. Third, socially desirable and trust-aware control designs for the CAVs are developed to improve the social and trustful traffic interactions between human drivers and CAVs. The human-centric CAVs for future transportation systems are pictured by effectively incorporating these three concepts. This dissertation is aimed to offer social and economical incentives such as increasing the efficiency of interactive traffic scenarios and boosting technology acceptance in today’s ground transportation sector.

Description

Keywords

mixed traffic, driver behavior modeling, model predictive control, connected and automated vehicles, human-automation interaction

Citation