Distributed intelligent control of street and highway ramp traffic signals

Abstract

The paper describes a distributed, knowledge-based system for real-time, traffic-adaptive control of traffic signals. The first of a two-stage learning process optimizes the control of steady-state traffic at a single intersection and over a network of streets. The second stage of learning deals with predictive/reactive control in responding to sudden changes in traffic patterns. The system can also display, store and retrieve information about traffic flow, intersection geometry, signalization, and performance of three- and four-leg intersections. In a design setup, it can compute the economically most advantageous geometry for an intersection, given the traffic flow and acceptable levels of service. The system has been tested in the laboratory on a range of scenarios (in terms of geometry and traffic flow) and has been found to perform as expected.