Adaptive traffic control at motorway bottlenecks with time-varying Fundamental Diagram

Abstract

This paper deals with the problem of controlling traffic at motorways bottlenecks in presence of an unknown, time-varying, Fundamental Diagram (FD). The FD may change over time due to traffic composition or to the presence of Connected and Automated Vehicles (CAVs) with varying driving characteristics and penetration rates. A novel methodology, based on Model Reference Adaptive Control, is presented to robustly estimate the time-varying set-points that maximise the bottleneck throughput. The proposed approach is integrated in a control scheme that includes a linear quadratic integral regulator designed to control traffic which comprises a percentage of CAVs. Simulation experiments, based on a first-order multi-lane macroscopic traffic flow model that also considers for the capacity drop phenomenon, are presented to illustrate the effectiveness of the proposed approach.