Urban road network traffic estimation method and system

Abstract

The invention relates to an urban road network traffic estimation method and system. With regard to the problem of coverage insufficiency of probe vehicle data on an urban road network, non-linear modeling is performed on urban road traffic correlation according to historical and real-time probe vehicle data, and real-time traffic speed estimation on all roads of the urban road network is achievedby parallel graph calculation. By the method and the system, the traffic estimation parallelism can be greatly improved, the calculation efficiency and the communication efficiency among calculationnodes are improved, near-real-time traffic estimation of the large-scale urban road network is achieved, and the method and the system have better practical application value.

Description

Technical field [0001] The invention belongs to the technical field of intelligent transportation, and specifically relates to a method and system for estimating urban road network traffic. Background technique [0002] With the rapid development of the national economy, the continuous acceleration of the urbanization process, and the continuous increase in the number of urban motor vehicles, the burden on urban transportation infrastructure continues to increase, and various transportation problems have become increasingly prominent. On the other hand, the rapid development and maturity of core technologies such as sensor technology, big data, and artificial intelligence have made the Internet of Things and smart cities gradually become the mainstream, greatly increasing people's expectations and demands for intelligent transportation. The realization of intelligent transportation is based on the comprehensive, accurate and real-time perception and estimation of the urban road n...

AI Technical Summary

Problems solved by technology

Such methods potentially linearly model the traffic correlation, although they can increase the calculation speed but cannot guarantee the accuracy of traffic estimation

Some researchers also proposed to use deep learning and other models (Lv, Y., Duan, Y., Kang, W., Li, Z., & Wang, F.Y. (2015). Traffic flow prediction with big data: A deep learning approach. IEEE Trans .Intelligent Transportation Systems, 16(2), 865-873.) Non-linear modeling of traffic correlation, but due to the high computational complexity of the nonlinear model itself, this type of method is only suitable for road networks containing a small number of road segments, and Large-scale traffic estimation for urban road networks containing tens of thousands of road segments is not possible

In general, the existing research at home and abroad still cannot well solve the city-wide traffic estimation based on the floating car perception data, and the following problems generally exist:

[0004] 1) Based on linear traffic correlation modeling, although it can reduce the computational complexity, it cannot guarantee the accuracy of traffic estimation

The actual urban traffic situation is more complicated, and the simple linear model cannot accurately describe the traffic correlation of the urban road network.

[0005] 2) Based on nonlinear traffic correlation modeling, although the accuracy of traffic estimation can be guaranteed, the existing methods have not designed a good mechanism to ensure the scalability of the system, and cannot be applied to large-scale traffic estimation within the city

Images