Malfunction characteristics of induction-loop sensors with freeway surveillance applications are addressed. Various types of malfunctions are identified, and their respective rates are computed on the basis of data from the I-880 surveillance system in the San Francisco Bay area. Specific algorithms are presented for diagnosing each malfunction type through inspection of aggregate sensor measurements. Also addressed are methods of repairing faulty measurement data, wherein estimates of the actual traffic condition are generated utilizing measurements from adjacent lanes.
Get full access to this article
View all access options for this article.
References
1.
PayneH. J., and ThompsonS. M.. Development and Testing of Incident Detection Algorithms: A Status Report on the Ongoing FHWA-Sponsored Project. Ball Aerospace & Technologies Corporation, Systems Engineering Operations, San Diego, Calif., 1995.
2.
PettyK.FSP 1.1—The Analysis Software for the FSP Project. University of California, Berkeley, 1994.
3.
PayneH. J., HelfenbeinE. D., and KnobelH. C.. Development and Testing of Incident Detection Algorithms, Vol. 2: Research Methodology and Detailed Results. Report FHWA-RD-76-20. FHWA, U.S. Department of Transportation, 1976.
4.
NihanN., JacobsonL. N., BenderJ. D., and DavisG.. Detector Data Validity. Washington State Department of Transportation, March1990.
5.
ChenL., and MayA. D.. The Use of Vehicle Detectors for Freeway Traffic Management, Vol. I: Detector Errors and Diagnosis. Working Paper UCB-ITS-WP-86-6, University of California, Berkeley, 1986.
6.
AndersonB. D. O. and MooreJ. B.. Optimal Filtering. Prentice-Hall, Inc., Englewood Cliffs, N.J., 1979.
