CEAE Faculty Candidate Speaker - Shi'an Wang: Harmonizing vehicle control and traffic priors in transportation cyber-physical systems (T-CPS): A unified framework

Abstract: Over the past four decades, there has been significant investment in intelligent transportation systems, with recent focus on connected and automated vehicles (CAVs). The introduction of CAVs offers a new paradigm for traffic estimation and control, with potential benefits such as increased traffic stability, reduced vehicle energy consumption and emissions, and improved urban mobility. However, mixed-autonomy systems, which include both CAVs and legacy vehicles, are complex and difficult to analyze and control. In this talk, I will present methods that integrate vehicle control and traffic priors for the analysis and optimization of emerging transportation cyber-physical systems (T-CPS)—systems that combine physical components, such as vehicles and infrastructure, with computational elements, such as sensors and software. Traffic instability often causes disruptions like shockwaves and capacity drops, which decrease traffic efficiency and increase energy consumption and emissions. These problems are most noticeable at bottlenecks, such as intersections and merges, but can also occur in the absence of these triggers, due to the collective behavior of human drivers. However, intelligently controlled CAVs can adjust their driving strategies based on real-time traffic conditions, helping to stabilize traffic, improve mobility, and increase efficiency. These benefits can also extend to legacy vehicles, due to the car-following behavior shared by all vehicles. By combining vehicle control with traffic flow theory, I will introduce a unified framework that allows for simultaneous traffic prediction and real-time control of CAVs, as well as the optimization of vehicle speed and powertrain. I will discuss the mathematical aspects of this framework and present results from simulations and real-world traffic data, showing up to a 16% reduction in energy use alongside improved traffic mobility. Finally, I will highlight the potential of T-CPS for creating more efficient, sustainable, and resilient cities in the future.