UNIST Presents AI System that Predicts Traffic Conditions

A recent study, affiliated with South Korea's Ulsan National Institute of Science and Technology (UNIST) has presented an artificial intelligence technology that can predict traffic conditions for the next 5 to 15 minutes at an error rate of less than four kilometers an hour.

From left are Professor Sungahn Ko and Chunggi Lee in the School of Electrical and Computer Engineering at UNIST.

A team of researchers, affiliated with UNIST presented an artificial intelligence technology that can predict traffic conditions for the next 5 to 15 minutes at an error rate of less than four kilometers an hour.

This breakthrough has been led by Professor Sungahn Ko and his research team in the School of Electrical and Computer Engineering at UNIST, together with Purdue University and Arizona State University. The new technology is currently being used by the Traffic Broadcasting Network (TBN) in Ulsan to provide traffic information to local residents. And it will soon to be supplied to cities, such as Gwangju, Busan, Daejeon, and Incheon within this year.

This interactive visual analytics system enables traffic congestion exploration, surveillance, and forecasting based on vehicle detector data. Through domain expert collaboration, we have extracted task requirements, incorporated the Long Short-Term Memory (LSTM) model for congestion forecasting, and designed a weighting method for detecting the causes of congestion and congestion propagation directions.

An overview of the AI-based traffic prediction system by Professor Sungahn Ko and his research team.

The new system is largely composed of two modules — One that analyzes and predicts the traffic situation and the other for visualizing the results.

Unlike previous traffic prediction systems that rely on probability and statistics to analyze past traffic records for prediction, the new system comes with higher accuracy by adding a deep learning algorithm that considers real-time traffic situations.

Indeed, it provides near real-time traffic estimation and prediction based on the real-time vehicle detector data to detect the causes of congestion and congestion propagation directions. Traffic situations predicted by the AI system are, then, visualized for easy understanding. Congestion levels and average driving speed, for instance, are described using colors and shapes.

“The new system learned past average speed in certain traffic locations along with congested areas on nearby roads, and traffic conditions during rush hour,” says Chunggi Lee in the Combined M.S/Ph.D. of Electrical and Computer Engineering at UNIST. “Using this system will allow navigation programs to notify the driver how current traffic conditions may change in the next 5 minutes.”

교통에 AI 적용…내비게이션 도착 정확해진다[뉴스8]

“The new data visualization technology will be implemented in the Urban Traffic Information Center (UTIC) website, so that anyone can easily understand the road traffic situation,” says Professor Ko. “This technology, which can utilize a large amount of traffic data, can also be used to find the optimal route during bad traffic situations, in conjunction with traffic broadcasting services or navigation programs.”

He adds, “In future work, we plan to perform more rigorous experiments by forecasting with other factors (e.g., weather, accidents) to develop more accurate forecasting models.”

The findings of this study had been achieved through cooperation with related experts, such as the National Police Agency, the Korea ROAD Traffic Authority (KoROAD), the Ulsan Traffic Broadcasting Network (TBN), and the Transportation and Construction Bureau of Ulsan Metropolitan City Hall.

The study on the new system has been published online in the journal, IEEE Transactions on Visualization and Computer Graphics.

Published: 03 Sep 2019

Contact details:

JooHyeon Heo

50 UNIST-gil, Ulju-gun, Ulsan, Republic of Korea, 44919

+82-52-217-1223
Country: 
Content type: 
Peer Reviewed