Field test of nonintrusive traffic detection technologies

Accurate, low-cost methods of collecting historical traffic information are essential in making well-informed transportation planning decisions. In addition, detection of real-time traffic conditions is a key element in advanced traffic management and traveler information systems. Until the last decade, inductive loop detectors, pneumatic road tubes, and temporary manual counts were the primary methods for collecting both real-time and historical traffic data. However, technological innovations have given rise to design many different types of advanced traffic detectors. Recently developed traffic detectors use sonic, ultrasonic, microwave, or infrared energy. Most of these detectors can be mounted overhead or to the side of traffic lanes. Magnetic sensors are now being built in sizes small enough to be placed in conduits under the roadway. Artificial intelligence algorithms can process videotaped images of road scenes and output many useful traffic parameters. Even though nonintrusive technologies have been available for several years, there are still many uncertainties regarding their use. Traffic engineers lack a comprehensive comparison of the various types of traffic detection technology. A study conducted by the Minnesota Department of Transportation (Mn/DOT) and SRF Consulting Group, Inc. (SRF) and sponsored by the Federal Highway Administration (FHWA) seeks to address this need. Mn/DOT and SRF undertook a two-year effort to test a wide variety of nonintrusive traffic detection technologies. The purpose of this evaluation was to collect practical information on the performance, installation requirements, long-term maintenance requirements, and costs of various types of nonintrusive traffic detection technologies. More than a dozen devices representing magnetic, sonic, ultrasonic, microwave, infrared, and video image processing technologies were evaluated during this project. Devices were evaluated for their performance in both freeway and urban intersection monitoring situations. Testing consisted of two phases. During Phase I, which ran from November 1995 to January 1996, all participating devices measured traffic data on three lanes of Interstate 394 in Minneapolis at the Penn Avenue interchange. Phase II, which ran from February to November 1996, consisted of an all-season monitoring of the devices' performance and maintenance requirements and involved both freeway and intersection installations. The Minneapolis-St. Paul metropolitan area provided an excellent opportunity to evaluate the devices in many types of weather extremes, including very cold and very hot temperatures, rain, snow, fog, and high winds.