March 27, 2018
Researcher Says Good Traffic Estimates Today Can Create Better Roads for Tomorrow
By Daniel CarrollMedia Inquiries
- College of Engineering
Carnegie Mellon University's Zhen (Sean) Qian, an assistant professor of civil and environmental engineering, is driving to make a smoother road for urban planners by creating more accurate ways to estimate traffic conditions.
The ability to estimate traffic — especially in urban environments — is important as it can allow for efficient, real-time traffic management, optimizing travel for emergency responders or the rerouting of traffic to adjust for delays. The data and estimates gathered also enable engineers like Qian to understand how the infrastructure can be better utilized, such as changing road markings or the timing of traffic lights.
Qian said in the past, the sheer number of roads and intersections in an urban environment had been too large to create accurate estimations, and it would be infeasible to place a sensor on every road and intersection. His approach, which uses a link queue model, takes advantage of large amounts of data from numerous pre-existing sources — smartphones, GPS devices, probe vehicles — in conjunction with strategically positioned sensors, and performs an efficient computation to output a reasonably good estimate.
Within a small portion of the Washington, D.C., area, Qian combined multiple data sets with two speed detectors to accurately estimate travel speed to an acceptable error rate within 8.5 percent.
"In the longterm, we want to know on average, throughout the entire year, and over multiple years, how traffic increases," Qian said. "This information can be used to inform engineers in designing the next generation of roadways and traffic infrastructure, shaping the street and highways of tomorrow."
Qian said the next goal is to apply the model over a larger area.
"The main challenge is ensuring the quality and coverage of our data," he said. "Mathematically, we're still determining if it's feasible to run such a model for a very large-scale urban network."
The Mobility Analytics Data Center (MAC), which Qian directs, already is developing a centralized data engine for compiling and computing large amounts of data sets from sources all throughout the city of Pittsburgh.
"The D.C. network for demonstration in this paper is small, but I'm still very excited about the progress that we're making," Qian said. "Now we're trying to find a real-world experiment here in Pittsburgh with the MAC center, where we already have all the data."
Qian and recent Ph.D. graduate Yiming Gu, were recently recognized for their work in this field. They were named the 2017 recipients of the Greenshields Prize for their paper "Traffic State Estimation for Urban Road Networks Using a Link Queue Model" by the Committee on Traffic Flow Theory and Characteristics of the Transportation Research Board. Gu is now a senior research engineer at the United Technologies Research Center.
Qian also was recognized by the National Science Foundation for his work in Pittsburgh. He has received a five-year, $500,000 NSF Faculty Early Career Development Award for his project "Probabilistic Network Flow Theory: Embracing Emerging Big Data for Efficient, Reliable and Sustainable Multi-modal Transportation Systems." This project will involve collaboration with several public agencies and private firms to develop, deploy and test real-world systems in the Pittsburgh metropolitan area based on large-scale data analytics.
Qian is a part of CMU's Metro21: Smart Cities Institute, a university-wide effort for research, development and deployment of solutions to improve the quality of life in communities. The institute addresses issues such as traffic congestion, pedestrian safety, road infrastructure, energy efficiency, law enforcement, health care, fire prevention, and air and water quality. Through its strengths at the nexus of technology and humanity, CMU is generating real-world solutions to the evolving needs of cities.