Control Barrier Function Based Speed Control for Fixed-Line Transit Headway Management
Abstract
We present a control barrier function (CBF) approach for headway management in fixed-line transit systems (FTLSs) through vehicular speed control. We frame headway management as a formal property in terms of vehicle time-differences. Using a dynamical model of vehicle speed control, we develop a combined CBF quadratic-programming (QP) supervisory approach. The CBF-QP supervises control speed inputs for forward-invariance of minimum and maximum time-differences between vehicles. While the need for vehicles to stop creates modeling error that can lead to property violations, the CBFs can recover back to satisfying the time-gap properties after violations. We present numerical experiments that compare the CBF-QP approach to other controllers. We find that the CBF-QP approach is able to supervise a poorly performing controller to achieve significantly improved headway regularity. We additionally compare the CBF-QP with an existing linear-quadratic control approach and find that the CBF-QP performs similarly or better.
Type
Venue
29th IEEE International Conference on Intelligent Transportation Systems (ITSC 2026)
Authors
Postdoctoral Researcher
George is a postdoctoral researcher studying formal methods (FM) approaches for real-time control of transit
systems. He is interested in using FM approaches to study how we can assure properties of transit operation while
also balancing the performance of surrounding traffic flows. George previously received his Ph.D. from the Vanderbilt
University Civil Engineering department, and his bachelor’s degree from the University of Illinois at
Urbana-Champaign Civil Engineering department.
Associate Professor
Aron Laszka is an Associate Professor in the College of Information Sciences and Technology at the Pennsylvania State University. Previously, he was an Assistant Professor at the University of Houston, a Research Assistant Professor at Vanderbilt University, and a Postdoctoral Scholar at the University of California, Berkeley.
His research interests revolve around artificial intelligence and cyber-physical systems, focusing on machine learning for decision making and optimization, with applications in smart and connected communities, critical infrastructure, and cybersecurity.
His research is funded by the National Science Foundation, the Department of Energy, the Department of Transportation, and other agencies.