Real-time Group Route Diversion Control Considering Dynamic Response Behavior

This paper proposes a real-time group route diversion control (GRDC) methodology to determine route diversion strategies based on the controller's objectives and drivers' dynamic response behavior. The GRDC optimization procedure is to update the GRDC strategies to minimize the difference between the desired diversion rates and the diversion rates obtained from the drivers' route diversion behavior (DRDB) model, which comprises a GRDC model that determines the desired diversion rates to satisfy some system objectives of the GRDC controller, and a controller-estimated drivers' route diversion behavior (DRDB) model that predicts the likely driver route choice decisions in light of the updated GRDC strategies. The experimental results show that the group route diversion control considering dynamic response behavior can effectively provide guidance for drivers, improve the accuracy of traffic state prediction, and obtain better system performance over the same period of operations.