An Online Learning Collaborative Method for Traffic Forecasting and Routing Optimization

Recent advances in technologies such as the Internet of Things (IoT) and Cyber-Physical Systems (CPS) have provided promising opportunities to solve problems in urban traffic. With the help of IoT technologies, online data from road segments are captured by monitoring devices, while real-time data from vehicles are collected through preinstalled sensors. Based on these data, a CPS model is constructed to depict real-time status and dynamic behavior of road segments and vehicles. An online learning data-driven model is developed to extract prior knowledge and enhance collaboration between road segments and vehicles by combining short-term traffic forecasting and real-time routing optimization. A case study based on Xi'an city is presented to demonstrate the feasibility and efficiency of the proposed method, showing a reduction in the travel time with reasonable computation time, without much compromising the travel distance and fuel consumption. This work potentially strengthens the transparency and intelligence of urban traffic systems.

Automated summary

Recent advancements in technologies like IoT and CPS have provided opportunities to address urban traffic issues. By utilizing these technologies, real-time data from road segments and vehicles are collected, and a CPS model is developed to depict dynamic behaviors. An online learning data-driven model is created to optimize traffic routes, demonstrating efficiency in reducing travel time and fuel consumption in a case study.