SYNOPSIS
North Carolina is preparing for CAVs through multiple initiatives including an NCDOT-sponsored effort “NC Readiness for Connected and Autonomous Vehicles (CAV)” and the Highly Automated Vehicle Committee appointed by the State Legislature. In providing a roadmap for safe deployment of CAVs on North Carolina roads, these initiatives highlight the need to consider modifications to roadway infrastructures. CAVs’ infrastructure needs can be broadly categorized either as improvements to existing traditional infrastructure or as emerging infrastructure needs. New infrastructure, including communications hardware to support Connected Vehicle (CV) applications, has been a recent focus of USDOT with three national pilot locations in Tampa, New York City, and statewide in Wyoming. In addition, NCDOT has joined the Signal Phase and Timing (SPaT) Challenge with a CV deployment on NC-55 in Cary.
This talk assesses the current NCDOT Programs that may be impacted by 1) the changes to the traditional infrastructure that are needed to safely deploy CAVs statewide, 2) expands on a low-latency edge computing architecture that can support CAV applications needing extensive computation or timely response, and 3) provides recommendations to NCDOT units and programs to increase statewide readiness for CAVs in both traditional and emerging infrastructure.
ABOUT THE SPEAKER
Mr. Chia-Hung Lin is currently a Ph.D. student under supervision of Dr. Shih-Chun Lin at the Department of Electrical and Computer Engineering, North Carolina State University. Mr. Lin has been a part of the NC-CAV Center research under Thrust 2 and is building next-generation vehicular communication systems and data-driven communication/networking algorithms to facilitate the development of CAV applications. Under Thrust 2, Mr. Lin has published a journal paper and two international conference papers. Specifically, he constructed a vehicular communication platform, which combines a realistic transportation module and a 5G specification-defined communication module, to develop and validate CAV algorithms. Using the platform, a spectral sensing algorithm and a cooperative neighboring vehicle positioning algorithm were developed, leveraging the power of deep learning technologies to improve the safety and efficiency of CAV applications.