Relying on a traffic simulator is often necessary when multiple traffic conditions need to be replicated, either to predict specific quantities of particular interest or to assess more general network properties, such as efficiency or resilience, under different scenarios. While potentially delivering a high level of fidelity, producing such a simulation may come at a prohibitive computational cost when it is applied in a real context, making this approach unsuitable for real-time applications in most cases. In this regard, the goal of the present work is twofold. Firstly, we aim to surrogate a traffic simulator with a data-driven approach in order to produce real-time traffic predictions that are also sufficiently accurate and effective. Secondly, we want to determine the minimum amount of data, i.e., the smallest number of sensors deployed on the network, that still allow to obtain predictions within a predefined bound. The effectiveness of the approach is evaluated on the full-scale urban network of Rapallo, Italy, in which we employ the AIMSUN NEXT simulator targeting the morning peak hours, i.e. between 7:00 a.m. and 9:00 a.m. In the paper, multiple state-of-the art ML algorithms are tested to assess their effectiveness as surrogate models under the considered problem.

The paper proposes a multi-class control scheme for freeway traffic networks. This control scheme combines two control strategies, i.e. ramp metering and route guidance, in order to reduce the total time spent and the total emissions in a balanced way. In particular, the ramp metering and route guidance controllers are feedback predictive controllers, i.e. they compute the control actions not only on the basis of the measured system state, but also on the basis of the prediction of the system evolution, in terms of traffic conditions and traffic emissions. Another important feature of the controllers is that they have a multi-class nature: different classes of vehicles are considered and specific control actions are computed for each class. Since the controllers are based on a set of parameters that need to be tuned, the overall control framework also includes a module to properly determine the gains of the controllers. The simulation analysis reported in the paper shows the effectiveness of the proposed control framework and, in particular, the possibility of implementing control policies that are specific for each vehicle type.

In this paper, a multi-class and multi-objective combined ramp metering and routing control strategy is proposed to improve performance in freeway traffic networks. The control strategy is of the multi-class type, i.e. different classes of vehicles are taken into account, so that specific control policies can be devised for each class. Moreover, ramp metering and route guidance are properly coordinated in order to reduce the total travel time and the total emissions in the freeway system in a balanced way. The controller proposed in the paper is of the predictive feedback type, i.e. the control computed at each time step depends on the measured system state and on the prediction of the system evolution, considering both a traffic and an emission model. Some simulation results are presented showing the effectiveness of the proposed control scheme.