Modelling and hierarchical optimization for oversaturated urban road traffic networks

Abstract

In this paper the problem of traffic control during the rush hour is tackled using optimal control theory. A simple discrete-time model is developed to describe the dynamic behaviour of oversaturated urban road traffic networks. It is seen that standard optimization techniques can not be used to obtain optimal control trajectories for this system due to the dimensionality difficulties which are particularly accentuated by the hard inequality constraints on the states and controls as well as the pure time delays which arise in the system description. However, a recently developed algorithm can be used to compute the global optimum in a computationally efficient way. This is demonstrated by simulation studies for two particularly complex junctions in London as well as for a small network which is a particularly well-known trouble spot in the London area.