Holistic control of smart mobility systems for efficiency, safety, and equity
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The rapid advancement of smart mobility systems, including connected and automated vehicles (CAVs), autonomous mobility-on-demand (AMoD) systems, and shared mobility, has the potential to transform urban transportation landscapes. These systems offer solutions to address the challenges of the increasing urban population, such as traffic congestion, pollution, and accidents. However, the successful implementation of these systems requires novel approaches that integrate decision-making across multiple levels, considering the complex interactions between vehicles, infrastructure, and societal factors. This dissertation focuses on three key areas: (1) traffic-aware routing for CAVs, (2) hierarchical frameworks for scalable routing and optimal control, and (3) achieving mobility equity through emerging mobility systems. ☐ The first contribution of this dissertation is the development of a traffic-aware routing framework for CAVs that integrates network-level routing with intersection-level coordination. By considering the actual motion profiles of each vehicle in the routing problem, the proposed approach captures the mutual impact of routing decisions and vehicle coordination, leading to more accurate and efficient routing solutions. The framework also includes a re-routing strategy to handle infeasible scenarios, ensuring the identification of person-by-person optimal solutions and allowing for the introduction of more vehicles into the network. ☐ The second contribution is the introduction of a comprehensive hierarchical decision-making framework that optimizes processes across multiple levels in smart mobility systems. The proposed framework integrates an upper-level flow-optimization problem for AMoD systems, a vehicle-level routing problem for CAVs, and a lower-level trajectory planning problem for individual vehicles at intersections. This integration enables the framework to adapt to infeasible solutions and provide optimal operational strategies across different levels of the transportation system. The dissertation also establishes the necessary optimality conditions for vehicle-level trajectory planning and proposes a heuristic approach for the practical modification of infeasible flow. ☐ The third contribution focuses on achieving mobility equity through the operation of emerging mobility systems. The dissertation introduces a novel Mobility Equity Metric (MEM) that evaluates the distribution of the ``ability to move" among different regions, considering various societal factors. A system planner is designed to improve the MEM within a multi-modal emerging mobility system by generating route suggestions that enhance mobility equity across different neighborhoods in a city. The effectiveness of the proposed MEM and routing approach is demonstrated using numerical simulations and real-world data, showcasing the potential of emerging mobility systems to improve mobility equity. ☐ Collectively, this dissertation contributes to the development of safer, more efficient, and equitable urban transportation systems in the era of smart mobility technologies. The multi-level decision-making frameworks presented in the dissertation address the complexities of smart mobility systems and provide optimal operational strategies that integrate coordination, routing, and mobility equity. The theoretical and practical insights gained from this research have the potential to guide the successful implementation of smart mobility systems in real-world scenarios, ultimately leading to improved transportation experiences for urban populations.
Description
Keywords
Connected and automated vehicles, Optimal control, Smart mobility systems, Autonomous mobility-on-demand