Model-Based Traffic Congestion Control in Urban Road Networks

Abstract

Traffic congestion is a serious problem for big cities. The model-based optimization control strategy is an effective method for decreasing traffic congestion. A proper control performance criterion is a key factor for effectively addressing different urban road network issues, for example, network mobility and traffic congestion. Therefore, two control performance criteria are proposed—L2-norm and L∞-norm performance indexes—for model-based urban road network controllers. The L2-norm and L∞-norm performance indexes are aimed at keeping road network homogeneity and decreasing road network congestion, respectively. For analysis of these two control performance criteria, case studies were conducted with the microscopic simulation software CORSIM to evaluate the designed model-based controllers with different criteria. The control effects of the model-based road network controllers were compared quantitatively by analyzing the degree of road network congestion and qualitatively by analyzing the macroscopic fundamental diagram of the road network (i.e., a network aggregated characteristic).