Skip to Main Content
The primary focus of this study is to present a new integrated traffic control model for urban freeways using model-based predictive control (MPC) framework, where the effect of traffic information on drivers' en-route diversion behaviour is explicitly modelled using an adaptive constrained Kalman filtering theory. This captures the effect of traffic information on drivers' real-time en-route diversion behaviour, based on on-line traffic surveillance data. The integrated control task is formulated as a dynamic, non-linear, discrete time optimal control problem with constrained control variables. The network traffic flow process is modelled by a dynamic network traffic flow model, which is deterministic, discrete in time and space, macroscopic and suitable for model-based traffic control. Feedback control is realised by solving the optimisation problem for each control interval over a long future-time horizon. Simulation results for a case study show that the proposed integrated MPC model takes in consideration of the time-dependent traffic characteristics and drivers' actual behaviour and can significantly enhance the traffic efficiency and reduce the cost of traffic system by 17.1-30.0%.