Hierarchical control for large-scale urban road traffic networks

Abstract

In the current work we focus on the development of hierarchical control structures to tackle the problem of signal control for large-scale urban networks. A recently developed perimeter control regulator, which integrates model-based optimal control and online data-driven learning/adaptation, is utilized for the upper-level layer. Another lower-level control layer utilizes the max-pressure regulator, which has been also proposed recently and constitutes a local feedback control law, applied in coupled intersections, in a distributed systems-of-systems (SoS) concept. Different approaches are discussed about the design of the hierarchical structure of SoS, i.e. mutual interactions between the two control layers, activation/deactivation of each layer, mutually related objectives of the regulators, online versus offline selection of critical intersection for the lower-level control layer. A hierarchical control approach that combines local and network level characteristics is expected to treat better uncertainties in demand and behavioural characteristics of drivers moving towards a more reliable performance of all users in the system.