Integrated Adaptive-Signal Dynamic-Speed Control of Signalized Arterials

Abstract

This paper demonstrates the feasibility of an advanced concept of traffic control for congested urban arterials in an intelligent transportation system (ITS) setting. The core idea of the new concept is to integrate the capability of dynamic adaptive signals with dynamically optimized time-dependent variable speed. Under such control, speeds would be automatically optimized and set by a central computer in parallel with the other signal control parameters. Speed and signal control parameters would change between links and over time, in response to changing traffic conditions. Drivers would follow the optimized speed as they enter a link. Once an optimal speed has been set for a link, it remains constant until the control cycle ends. Link speeds would be updated only at the end of every control cycle. The control cycle may change in length as system conditions evolve. The new control concept was tested on a congested arterial with multiple links. The arterial system was modeled as a discrete event time varying dynamic system with a control period spanning several cycles. System throughput was maximized subject to such critical operational measures as intersection blockage, queue spillbacks, and other relevant traffic operation measures. Genetic algorithms were used as an optimization tool. Results show that the proposed control concept will significantly improve traffic flow. The new control concept is suitable for on-line implementation in an ITS setting.