Model-based control of intelligent traffic networks

Road traffic networks are increasingly being equipped and enhanced with various sensing, communication, and control units, resulting in an increased intelligence in the network and offering additional handles for control. In this chapter we discuss some advanced model-based control methods for intelligent traffic networks. In particular, we consider model predictive control (MPC) of integrated freeway and urban traffic networks. We present the basic principles of MPC for traffic control including prediction models, control objectives, and constraints. The proposed MPC control approach is modular, allowing the easy substitution of prediction models and the addition of extra control measures or the extension of the network. Moreover, it can be used to obtain a balanced trade-off between various objectives such as throughput, emissions, noise, fuel consumption, etc. Moreover, MPC also allows the integration and network-wide coordination of various traffic control measures such as traffic signals, speed limits, ramp metering, lane closures, etc. We illustrate the MPC approach for traffic control with two case studies. The first case study involves control of a freeway stretch with a balanced trade-off between total time spent, fuel consumption, and emissions as control objective. The second case study has a more complex layout and involves control of a mixed urban-freeway network with total time spent as control objective. B. De Schutter Delft Center for Systems and Control & Marine and Transport Technology, Delft University of Technology, e-mail: [email protected] H. Hellendoorn, M. van den Berg, S.K. Zegeye Delft Center for Systems and Control, Delft University of Technology, e-mail: {j.hellendoorn,monique.vandenberg,s.k.zegeye}@tudelft.nl A. Hegyi Department of Transport & Planning, Faculty of Civil Engineering and Geosciences, Delft University of Technology, e-mail: [email protected]