A comparative analysis of solution algorithms for nonlinear freeway traffic control problems

Objective of this paper is to test and compare different solution algorithms for solving nonlinear freeway traffic control problems. The motivation of this work is related to the fact that the nonlinear problems arising in freeway traffic control are very challenging from a computational point of view, since they are generally characterized by a large number of variables and are often nonlinear and nonconvex. Hence, efficient solution algorithms for solving these problems are required in order to design efficient control frameworks to be applied in large freeway networks, possibly in real time. In this paper, some solution algorithms, either gradient-based or derivative-free, are tested on a specific freeway traffic control problem, i.e. the definition of a two-class ramp metering strategy to reduce congestions and emissions in a freeway.