Reducing the Effect of Inaccurate Lane Identification on PP69-10-Based Rut Characterization
Publication: Journal of Infrastructure Systems
Volume 22, Issue 1
Abstract
The new draft AASHTO standard practice PP70-10 and PP69-10 are promising to change the landscape of rut measurement. PP70-10 specifies the technical requirements for transverse profiling and PP69-10 defines multiple parameters such as percent deformation and rut cross-sectional area to characterize a rut. Per PP70-10, an inspection device should cover at least 4.0 m (13 ft) transverse width in order to cover the full lane under vehicle wandering. However, for many existing transverse profilers, it is challenging to identify the lane markings from the raw images in a consistent manner. This deficiency would prevent these systems from accurately calculating PP69-10-based rut attributes. In this study, with the field data collected in Arkansas national highway systems (NHS) using an automated system called PaveVision3D Ultra, the impact of lane offsets on PP69-10-based rut attributes is investigated. The rut attributes calculated with programs embedded in the system, where accurate lane locations are semiautomated extracted from both 2D and 3D images, are set up as base values. It is found that lane offsets greater than 50 mm would cause significant errors in rut measurement. To reduce the negative effect of inaccurate lane identification on rut characterization, a methodical framework is proposed in this research. A case study is performed to demonstrate that machine learning approaches could satisfactorily aid in increasing the accuracy of rut characterization even when the locations of actual lane markings are unknown. This research would improve the reliability and accuracy of the new AASHTO rut measurement.
Get full access to this article
View all available purchase options and get full access to this article.
Acknowledgments
The authors would like to thank WayLink System. and OSU students who assist in data collection, system calibration, and data preprocessing. Thanks also go to the anonymous reviewers who helped improve the organization of this paper.
References
AASHTO. (2013a). “Standard practice for collecting the transverse pavement profile.”, Washington, DC.
AASHTO. (2013b). “Standard practice for determining pavement deformation parameters and cross slope from collected transverse profiles.”, Washington, DC.
Antunes, M., et al. (2008). “COST action 354: Performance indicators for road pavements.”, FSV-Austrian Transport Research Association, Vienna, Austria.
Archilla, A. (2006). “Repeated measurement data analysis in pavement deterioration modelling.” J. Infrastruct. Syst., 163–173.
ASTM. (2011). “Standard test method for measuring rut-depth of pavement surfaces using a straightedge.”, West Conshohocken, PA.
Breiman, L. (2001). “Random forests.” Mach. Learn., 45(1), 5–32.
Breiman, L., Friedman, J., Stone, C., and Olshen, R. A. (1984). Classification and regression trees, Champman & Hall/CRC, Boca Raton, FL.
Chen, D., and Li, Z. (2008). “Comparisons of five computational methods for transverse profiles.” J. Test. Eval., 36(5). 1–8.
Conway, D., and White, J. (2012). Machine learning for hackers, O’Reilly Media, Sebastopol, CA.
FHWA (Federal Highway Administration). (2010). “Highway performance monitoring system field manual.” Washington, DC.
FHWA (Federal Highway Administration). (2012). “Manual on uniform traffic control devices for streets and highways.” Washington, DC.
FHWA (Federal Highway Administration). (2013). “MAP-21: Moving ahead for progress in the 21st century.” 〈www.fhwa.dot.gov/map21/〉 (Aug. 8, 2013).
Haas, R., Hudson, W. R., and Zaniewski, J. P. (1994). Modern pavement management, Krieger Publishing, Malabar, FL.
Huang, Y., Copenhaver, T., and Hempel, P. (2013). “Texas department of transportation 3D transverse profiling system for high-speed rut measurement.” J. Infrastruct. Syst., 221–230.
Huang, Y. H. (2001). Pavement analysis and design, 2nd Ed., Prentice-Hall, Englewood Cliffs, NJ.
Johnson, D. (1998). Applied multivariate methods for data analysis, Duxbury Press, Grove, CA.
Lenngren, C. A. (1988). “Some approaches in treating automatically collected data on rutting.”, Transportation Research Board, Washington, DC, 20–26.
Liaw, A., and Wiener, M. (2002). “Classification and regression by randomforest.” R News, 2(3), 18–22.
McGhee, K. H. (2004). “Automated pavement distress collection techniques: A synthesis of highway practice.”, Transportation Research Board, Washington, DC.
R Development Core Team. (2012). “R: A language and environment for statistical computing.”, Vienna, Austria.
Rumelhart, D. E., McClelland, J. L., and The PDP Research Group. (1986). Parallel distributed processing: Explorations in the microstructure of cognition, Vols. 1 and 2, MIT Press, Cambridge, MA.
Simpson, A. (2001a). “Characterization of transverse profiles.”, Fugro-BRE, Austin, TX.
Simpson, A. (2001b). “Measurement of rut in asphalt pavements.” Ph.D. dissertation, Univ. of Texas, Austin, TX.
Svozil, D., Kvasnicka, V., and Pospichal, J. (1997). “Tutorial: Introduction to multi-layer feed-forward neural networks.” Chemom. Intell. Lab. Syst., 39(1), 43–62.
Tsai, Y., Wang, Z., and Li, F. (2011). “Assessment of rut depth measurement using emerging 3D continuous laser profiling technology.” Transportation Research Board 90th Annual Meeting, Transportation Research Board, Washington, DC.
Venables, W., and Ripley, B. (2002). Modern applied statistics with S, 4th Ed., Springer, New York.
Wang, K. (2013). “Rapid network level survey with automated cracking detection.”, Arkansas State Highway and Transportation Dept., Little Rock, AR.
Wu, X. (2012). Statistical methods for complicate data sets—Applications based on R, China Renmin University Press, Beijing (in Chinese).
Information & Authors
Information
Published In
Copyright
© 2015 American Society of Civil Engineers.
History
Received: Mar 25, 2014
Accepted: May 5, 2015
Published online: Jul 6, 2015
Discussion open until: Dec 6, 2015
Published in print: Mar 1, 2016
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.