Interested in CR2C2 activities and engagement opportunities?
Over a quarter of all road fatalities are curve related. This is a high-priority societal challenge in the US. The MUTCD (Manual on Uniform Traffic Control Devices, FHWA, 2012) requires various horizontal alignment warning signs (curve signs) and adequate advisory speed to ensure curved roadway safety. However, the majority of local transportation agencies (counties and cities) have are yet to meet the MUTCD requirements. In addition, the current practice for assessing the MUTCD compliance of existing curve warning signs requires a lot of effort to manually inventory existing signs, and measure and verify their sign type, placement, and spacing. Therefore, there is an urgent need for a low-cost solution since the majority of local transportation agencies with limited resources cannot afford the current practice. This talk will present a cost-effective curve safety assessment methodology and technology application using smart phones and Artificial Intelligence (AI) technologies developed through a competitively selected research project sponsored by the National Academy of Science (NAS) National Cooperative Highway Research Innovation Deserving Exploratory Analysis (IDEA) program and the Georgia Department of Transportation (GDOT). A cost-effective method has been developed for automatic curve sign design and MUTCD-compliant checking using low-cost mobile devices, AI and crowdsourcing technologies with a test performed on 26 miles of State Route 2 in GDOT District 1. The developed technology can also help transportation agencies identify and prioritize the roadways for safety improvements, like High Friction Surface Treatment (HFST) with benefit-cost analysis.
Yichang (James) Tsai is a professor in the School of Civil and Environmental Engineering and an adjunct professor in the School of Computer and Electrical Engineering (CEE) at Georgia Tech. He is currently the group leader of Construction and Infrastructure Systems Engineering (CISE) in CEE at Georgia Tech. Dr. Tsai’s research focuses on applying sensing technologies (3D laser, Lidar, and smartphone technologies), computer vision, AI, and GIS spatial analysis to 1) automated pavement condition evaluation and asset management, 2) transportation safety, 3) vehicle energy-emission reduction, and 4) safe mobility of the aging population. He has developed and successfully implemented the complex, large-scale, GIS-based, Risk-based Georgia Pavement Management System (GPAMS) for the Georgia Department of Transportation (GDOT). GDOT has used this system to assess, preserve, and manage its 18,000 centerline miles of highway over the past 20 years. Dr. Tsai’s research project received the 2017 AASHTO High-Value Research Award, a national award in the US, recognizing its innovation and successful implementation of an automatic pavement condition evaluation method using 3D laser and AI technologies. Since 2010, he has served as the Associate Editor of ASCE’s Journal of Computing in Civil Engineering.
