01517586

Component

https://doi.org/10.3141/2443-07

http://scholar.google.com...etty&publication_year=2014

This paper describes the development of a hybrid reidentification algorithm to estimate travel times entirely on the basis of data that are readily available from single loop detectors (60-Hz vehicle occupancy data) without the need for additional hardware in the field. The key concept is data fusion, which combines crude and potentially inaccurate spot-point estimates with a software-based signature-matching algorithm. The method was applied to real data from an urban freeway site with four locations and six segments, with Bluetooth-matched travel time as the ground truth. The method performed satisfactorily, with a relative error of 9% for a segment 1.54 mi long, among others. It removed bias and improved the baseline spot-speed method significantly. The approach requires minimum calibration with no additional hardware. The method, therefore, proves suitable for widespread deployment and provides a clear path for agencies to leverage their existing loop and controller infrastructure for accurate travel time estimation through reidentification.

01548337

14-3101

English

pp 56–62

2014

Transportation Research Record: Journal of the Transportation Research Board

9780309295314

Print

Figures (6) ; References (33) ; Tables (2)

Algorithms; Automatic vehicle identification; Bluetooth technology; Data fusion; Estimating; Loop detectors; Travel time

Data and Information Technology; Highways; Operations and Traffic Management; I72: Traffic and Transport Planning

TRIS, TRB, ATRI

Jan 27 2014 3:04PM

• Algorithm for Detector-Error Screening on Basis of Temporal and Spatial Information
• Application of Seasonal Adjustment Factors to Subsequent Year Data
• Automated Analysis of Pedestrians’ Nonconforming Behavior and Data Collection at an Urban Crossing
• Effect of Traffic Load Input Level on Mechanistic–Empirical Pavement Design
• Estimating Spatially and Temporally Continuous Bicycle Volumes by Using Sparse Data
• Estimation of Annual Average Daily Bicycle Traffic with Adjustment Factors
• Exploring Bias in Traffic Data Aggregation Resulting from Transition of Traffic States
• Hybrid Prediction Approach Based on Weekly Similarities of Traffic Flow for Different Temporal Scales
• Identification of Network Sensor Locations for Estimation of Traffic Flow
• Institutionalizing Bicycle and Pedestrian Monitoring Programs in Three States
• Process to Determine Locations for Motorcycle Counts
• Reidentification of Trucks on Basis of Axle-Spacing Measurements to Facilitate Analysis of Weigh-in-Motion Accuracy
• Semiautomated Tool for Extraction of Microlevel Traffic Data from Videographic Survey
• Vehicle Detector Evaluation Based on Concepts of Traceability and Confidence Interval
• Vehicle Length Measurement and Length-Based Vehicle Classification in Congested Freeway Traffic