Leveraging Sensor-Based Vehicle Position and Speed Information in Traffic Signal Control with Existing Infrastructure

Abstract

In recent years, many advanced control concepts for traffic signals have been proposed that make use of enhanced information about vehicle speeds and positions. Such information is anticipated to be yielded by connected vehicles, which will require time to become widely available. Similar albeit imperfect information can be extracted from existing detection systems, but most signal controllers are unable to directly access that information. The current paper examines preliminary results from a system that integrates vehicle speed and position data from a radar detection system directly into the actuated control logic, using an algorithm that seeks to minimize impacts of dilemma zones as well as reduce stops. Preliminary results from field implementation at three intersections yielded a 98% reduction in dilemma zone vehicles and a 38% reduction in stops, based on visual observation of traffic. The paper concludes by presenting plans for future development of the control concept.