Optimal routing for intelligent vehicle highway systems using a macroscopic traffic flow model

We consider Intelligent vehicle highway systems (IVHS) consisting of automated highway systems on which intelligent vehicles organized in platoons drive to their destination, controlled by a hierarchical control framework. In this framework there are roadside controllers that manage single stretches of highways. A collection of highways is then supervised by so-called area controllers. We focus on the optimal route choice control problem for the area controllers. In general, this problem is a nonlinear integer optimization problem with high computational requirements, which makes the problem intractable in practice. Therefore, we first propose a simplified but fast simulation model to describe the flows of platoons in the network. This model is a modified version of the macroscopic METANET traffic flow model, adapted to the case of platoons. Next, we use this model in a model-based predictive control approach in order to determine optimal splitting rates at the network nodes. These splitting rates can subsequently be communicated to the roadside controllers, which translate them into actual route instructions for the individual platoons.