A Traffic Flow Model and Intelligent Control Technique for Urban Trunk Road

Shen Guojiang,Sun Youxian
DOI: https://doi.org/10.1109/wcica.2004.1343741
2004-01-01
Abstract:This paper uses large-scale systems decomposition-coordination principle, fuzzy theory and neural networks technique to solve the problem of real time coordinated control for urban trunk road. Firstly, a neural macroscopic dynamic model based on the Kashani model for urban trunk road is proposed. Secondly, a two-level coordinated fuzzy control method implemented by neural network is presented. Based on the traffic volume data measured from each intersection the coordinated layer is coordinated with the number vehicle between the adjacent intersections. The operating layer adjusts on-line signal cycle and splits of every intersection. The object of the control method is to unblock the traffic trunk road and to shorten the average vehicle delay time. The simulation shows the proposed method has better performance than the fixed time control method.
What problem does this paper attempt to address?