Smart and Scalable Urban Traffic Control
[Dr. Xiao-Feng Xie, the lead inventor of SurTrac ]
Smart and Scalable Urban Traffic Control is a real-time adaptive traffic control system, which combines artificial intelligence (AI) and traffic theory to optimize highly dynamic traffic flow in complex real-world urban road networks.
His relevant research work also includes: multimodal traffic control (assisted with machine learning and computer vision techniques), integration with decentralized route choice models and dynamic congestion pricing protocols, vehicle-to-infrastructure (V2I) communication with connected vehicles, energy efficiency optimization, and data-driven self-learning and active congestion management based on performance measurement.
- (East Liberty) Deployed the the initial smart traffic lights for a network with nine intersections in the East Liberty neighborhood, Pittsburgh, PA. The system has been running since June 2012. Improvements of over 25% reductions in travel times, over 40% reductions in idle time, and a projected reduction in emissions of over 21% were achieved.
- (Bakery Square) Expanded the intelligent traffic signal control system further to include nine more intersections in the Bakery Square neighborhood, Pittsburgh, PA. The expanded system has been running since Oct 2013. Improvements of over 24% reductions in travel times, over 41% reductions in idle time, and a projected reduction in emissions of over 20% were achieved.
Disclaimer: Strong domain expertise is required to configure the smart and scalable urban traffic control system for achieving its optimal operations and improving urban traffic in the real world. Dr. Xie takes no responsibility for, and will not be liable for, any problems caused by inappropriate configuration of this system, for any installation or adjustment without his participation or guidance.
Xiao-Feng Xie, S. Smith, G. Barlow. Coordinated look-ahead scheduling for real-time traffic signal control. International Joint Conference on Autonomous Agents and Multiagent Systems (AAMAS), Valencia, Spain, 2012: 1271-1272. [DOI]