Trustworthy AI for Traffic Operation and Control

The rapid integration of artificial intelligence (AI) into transportation systems holds great potential for transforming urban mobility, especially with the explosion of transportation data and the growing capabilities of autonomous vehicles. Unlike conventional strategies, AI-based traffic sensing and control leverage massive historical and real-time data collected from diverse sensors to train learning-based algorithms for accurate state estimation, prediction, and adaptive decision-making. These algorithms are well-suited to handle heterogeneous transportation data, interpret human-dominated environments, and make informed decisions, ultimately helping alleviate congestion in complex, dynamic urban traffic systems.

Despite these technological advances, the practical adoption of AI in transportation systems has lagged, largely due to concerns about trustworthiness. Specifically, the inherent "black-box" nature of many AI algorithms introduces substantial challenges related to safety assurance, robustness and data privacy. These challenges directly impact public trust and system reliability, posing significant barriers to the widespread implementation of AI in transportation systems.

This workshop aims to bridge gaps in the design, deployment, and regulation of AI for traffic sensing and control. The program features invited talks from leading researchers and practitioners, offering a focused venue to share emerging results, practical lessons, and perspectives on trustworthy AI in transportation.