Coordinate Descent based Decomposed Anticipatory Dynamic Traffic Control

Abstract

Network Traffic Control has the potential for significantly reducing congestion, while avoiding excessive costs involved with infrastructural interventions.

Traffic networks inherently exhibit several degrees of decomposition, both hierarchically and geographically, and traffic control has been, in practice, severely influenced by this decomposition.

In this paper, we first deal with determining the conditions (if any) under which optimality can be preserved when distributing Static Anticipatory Optimal Control, and subsequently we exploit this knowledge to develop a better coordination mechanism in the Dynamic Anticipatory Traffic Control domain.