Early indications from the use of the newest vehicle detectors for high-speed signalized intersections suggested that they perform well as replacements for the Texas Department of Transportation’s (TxDOT’s) legacy systems, but this early conclusion needed verification based on rigorous field testing in a variety of traffic and environmental conditions. This research investigated the performance characteristics of detectors designed for the stop line area and indecision zone detection. In some cases, new detectors involved two technologies to cover both upstream and stop line areas. Increasing use of infrared (IR) cameras with video imaging systems was an attempt to overcome some of the limitations of traditional video detection. While these IR cameras may improve video detection for some lighting and temperature conditions, evidence suggested that they do not improve detection performance under all conditions. The traditional detection method that has been used by TxDOT on high-speed signalized intersection approaches for many years involved multiple detection points with inductive loops being the early favorite in terms of technology. However, TxDOT districts began adopting video imaging systems to replace loops as video began to show sufficient improvement, even though they were not as accurate as loops. This video trend continued to the point that the usage of video surpassed the usage of loops. However, the subsequent availability of other technologies at a reasonable cost caused TxDOT to seriously consider replacing both loops and video with newer systems that were immune to weather and lighting issues. Most of the newer systems also overcame the challenge of traffic interference and weakening pavement that plagued loops. The objectives of this research were to: Determine current TxDOT-specific needs for new vehicle detectors. Identify the most promising detectors for both stop line and dilemma zone detection. Develop guidelines on each new technology and establish recommended settings to guide TxDOT on installation and use of each detector and combination of detectors. In Task 1 Develop Test Plan, the research team proposed a list of detectors, general locations for tests, and the methodology of testing for consideration by TxDOT for this research project. The plan involved high-speed tests first in the controlled environment of the Texas A&M University (TAMU) Riverside campus followed by installation and testing of selected detectors at TxDOT signalized intersections. The research team installed detectors at Riverside as soon as the detectors became available and completed their installation in January 2015. Given the fact that most of the detectors selected for inclusion in this research were either new or relatively new and not well known, several discoveries in the early testing required modifications to the work plan. Two of the issues that required a change to the methodology involved radar detectors either for upstream detection or stop line detection. The initial issue involved operating multiple radar detectors simultaneously in close proximity to each other, resulting in possible interference. The research team could not determine conclusively based on resulting data nor could the consulted experts determine whether interference would occur for sure. Therefore, testing beyond that point in time occurred with only one system running at a time with the exception of the two detectors from Wavetronix, which were designed to be operated together. The other radar issue involved one of the newest radar detectors using a process similar to an existing radar detector whose process was patented. Both of these detectors were included in the project. As testing progressed at Riverside, the manufacturer of the newer detector had to stop manufacturing its product with the conflicting firmware. Fortunately, that manufacturer had built in a secondary method to replace its primary method. All of the data that the research team had collected to that point would not apply to the alternate procedure, so researchers had to start over for that detector. Another change in the work plan during the Riverside tests resulted from difficulties with one of the hybrid detectors selected in the data collection plan. Several challenges in early attempts to make it detect properly indicated that its continued testing in this project was premature. The research team worked with the manufacturer through both hardware and firmware changes to achieve proper performance, but the effort became more than could be sustained and meet project goals. This research report consists of six chapters organized by topic. Chapter 2 provides the results of a thorough literature review. Chapter 3 describes the efforts involved in conducting field data collection under controlled conditions at the Texas A&M University Riverside campus. Chapter 4 follows the controlled tests with intersection tests at four intersections in the TxDOT Austin and Houston Districts. Chapter 5 presents results and analysis of the field data as a prelude to developing the Guidelines for New Vehicle Detectors in Chapter 6. A component of the Guidelines is a Quick Reference Guide on strengths/weaknesses of various detectors included in this research. (Author/publisher)
Abstract