iTensor: intelligent traffic prediction and controls for smart city

A demo of intelligent signa control and hardware is here

The fast development of computing, sensor, information and communication technologies has stimulated the innovations and applications in Intelligent Transport Systems (ITS) technologies and associated industrial innovations on vehicles and infrastrures. For management and operations of road traffic, there is big potential to make the system more intelligent, energy-efficient and eco-friendly. During the past decade, TrafficLab has been developing ITS solutions for modern traffic information and management systems in the context of smart city.

iTensor started due to an industrial project for developing innovative traffic control technologies in 2011. Then it turned out to be a doctoral research project of developing agent-based traffic control framework using both model-free and model based reinforcement learning technologies in 2013. The methodological approaches established in the doctoral thesis of Junchen Jin (2018) show big potential to improve traffic signal control technologies for real infrastructure systems. The computational algorithms were implemented even in real signal controllers.  The project got both several public and private fundings and led to several important publiations in top journals of the field including

J. Jin and X. Ma, "A group-based traffic signal control with adaptive learning ability," Engineering Applications of Artificial Intelligence, vol. 65, pp. 282-293, 2017.

J. Jin and X. Ma, "Hierarchical multi-agent control of traffic lights based on collective learning," Engineering Applications of Artificial Intelligence, vol. 68, pp. 236-248, 2018.

J. Jin and X. Ma, "A Multi-objective agent-based control approach with application in intelligent traffic signal system," IEEE Transactions on Intelligent Transportation Systems, Vol. 20, No. 10, pp. 3900-3912, 2019.

J. Jin, D. Rong, F. Zhu, H. Guo, X. Ma, and F. Wang, "PRECOM: A parallel recommendation engine for control, operations, and management on congested urban traffic networks" , IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 7, pp. 7332-7342, July 2022.

In order to support intelligent traffic signal system, iTensor also developed estimation technologies for real-time traffic states and prediction algorithms since 2016. The research has led to several later publications:

J. Jin and X. Ma, "A non-parametric Bayesian framework for traffic-state estimation at signalized intersections," Information Sciences, vol. 498, pp. 21-40, 2019.

M. Sederlin, X. Ma and J. Jin, "A hybrid modelling approach for traffic state estimation at signalized intersections", the 24th IEEE Conference on Intelligent Transportation Systems, 2021. 

J. Jin, D. Rong, T. Zhang, Q. Ji, H. Guo, Y. Lv, X. Ma, F. Wang,  A GAN-based short-term link traffic prediction approach for urban road networks under a parallel learning framework, IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 9, pp. 16185-16196, Sept. 2022.

Q. Ji, J. Jin, Y. Qin, X. Ma, Y. Zhang, GraphPro: A Graph-Based Proactive Prediction Approach for Link Speeds on Signalized Urban Traffic Network,  the 2022 IEEE International Conference on System, Man and Cybernetics (IEEE SMC2022), 2022.