Arterial Travel Time Estimation Using SCATS Detectors
ruey long cheu,derhorng lee,qun liu
DOI: https://doi.org/10.1061/40632(245)5
2002-01-01
Abstract:This paper introduces a model to estimate average link travel time of signalized arterials using data obtained from detectors in Sydney Co-Ordinated Adaptive Traffic System (SCATS). The proposed model makes use of detectors at the upstream stop line of a link to capture the arrival of vehicles, and use a platoon dispersion model to project the vehicle arrival times at the downstream stop line. Using the vehicle departure pattems obtained at detectors at the downstream stop line, the average link travel time is then estimated. The accuracy of the proposed model has been tested, by means of microscopic traffic simulations, for link lengths between 300 m to 800 m, and at different urban street level of service. Results indicate that the proposed link travel time estimation model has an average absolute error of 2.26 sec. INTRODUCTION With the development of Advanced Traveler Information Systems (ATIS), it has become more important to obtain accurate real-time travel time data that reflects the operating efficiency of a traffic network, for assessing the performance of traffic management strategies, and for provision of route-guidance information. Several researches have been conducted on arterial travel time or speed estimation despite the complexity caused by traffic signals at intersections. Most of these arterial travel time or speed estimation models rely on data gathered by loop detectors at a distance away from the stop line (Gault 1981, Sisiopiku and Rouphail 1994, Zhang 1999 and Xie et al. 2001). Relatively few efforts have been made for the case with detectors installed at the stop line, with the exception of Luk and Cahill (1986), because it is difficult to make use of stop-line detectors to measure the representative link speed, vehicle arrival rate and queue length at an intersection approach. In addition, conventional way of aggregating loop detector data at stop line at cyclic interval does not provide data of sufficient granularity. With the improvement and cost reduction in loop detector hardware, it is becoming possible to program new data processing logic into a detector's controller card, to record vehicle arrivals and departures over the detector at intervals of several seconds. It is with this in mind that a method of using stop-line detectors to estimate arterial travel time is developed and tested in this paper. The proposed arterial travel time estimation method makes use of the existing stop-line detectors of the traffic signal control system in Singapore named Green Link Determination, or GLIDE in short (Chin 1993). It is the local version of the wellknown Sydney Co-Ordinated Adaptive Traffic Systems (SCATS) (Sims and Assoc. Prof., Dept. of Civ. Engrg., Nat. Univ. of Singapore, 1 Engineering Dr. 2, Singapore 117576, Singapore (corresponding author). Email: cheu@nus.edu.sg. 2 Res. Scholar, Dept. of Civ. Engrg., Nat. Univ. of Singapore. 3 Asst. Prof., Dept. of Civ. lEngrg., Nat. Univ. of Singapore.