01155641

Component

Loop detectors are the most commonly used vehicle detectors for freeway management systems. There has been considerable research to screen the quality of loop detector data using macroscopic measurements (over 30 sec to five min samples) and more recently using microscopic data (i.e., individual vehicle actuations). However, some significant detector errors have not received much attention due to the difficulty of identifying their occurrence. This paper examines one such error, splashover, the erroneous detection in one lane of a vehicle from an adjacent lane. We examine the nature of splashover and develop an algorithm to find detectors exhibiting chronic splashover problems. Loop detector data with concurrent video-recorded ground truth data are used for the evaluation. The developed algorithm estimates the percentage of pulses in the subject lane that arise from vehicles in the adjacent lanes. In the algorithm, whenever one pulse in the subject loop is bounded by a pulse from an adjacent lane it is considered as suspected splashover. Care is taken to control for the fact that concurrent actuations will also arise from two vehicles passing the station in adjacent lanes by calculating a background reference. Since an extreme difference of loop detector sensitivity in adjacent lanes can impede the performance of the algorithm, the daily median on-time is used to calibrate the loop detector sensitivity. The algorithm was tested over several detector stations with and without splashover, and it exhibited good performance.

01147878

10-1654

English

21p

2010

Transportation Research Board 89th Annual Meeting

DVD

Figures (5) ; References (14) ; Tables (3)

Algorithms; Automatic vehicle detection and identification systems; Data quality; Error analysis; Lane occupancy; Loop detectors

Data and Information Technology; Highways

Transportation Research Board Annual Meeting 2010 Paper #10-1654

BTRIS, TRIS, TRB

Jan 25 2010 10:44AM