Loop Detectors for Vehicle Classification, Idea #32: Literature Search Thursday, July 14, 2016 By Jim Byerly, MLIS

MnDOT Library Collection In-situ vehicle classification using an ILD and a magnetoresistive sensor array / prepared by Stanley G. Burns. Publisher Minneapolis, Minn. : Intelligent Transportation Systems Institute, Center for Transportation Studies, University of Minnesota, [2009] Physical Details 1 v. (various pagings) : ill. (some col.) ; 28 cm. Series ( CTS ; 09-06) General Note "CTS 09-06"--Tech. report doc. p. "February 2009." Bibliography Includes bibliographical references. Report Note Final report. Summary This report provides a summary of results from a multi-year study that includes both the use of inductive loop detectors (ILDs) and magnetoresistive sensors for in-situ vehicle classification. There were strengths and weaknesses noted in both type of sensor systems. Although the magnetoresistive array provides the best vehicle profile resolution, the standard inductive loop detector provides a significant cost, hardware and software complexity, and reliability advantage. The ILD installed base far exceeds the number of magnetoresistive sensors. Several electrical and computer engineering students participated in the study and their contributions are included in the individual chapter headings. Under my direction, these students also presented project work and Research Day conferences at MN/DOT District 1 Headquarters. Addnl Physical Form Also available online via Internet. Funding Performed by University of Minnesota Duluth, Dept. of Electrical and Computer Engineering, sponsored by the Intelligent Transportation Systems Institute, Center for Transportation Studies, University of Minnesota CTS project #2002042 #2003044 #2004021 #2005012 Local Note CTS-09-06 Electronic Version http://www.cts.umn.edu/Publications/ResearchReports/reportdetail.html?id=1750 Browse Call Number TE228 .B78 2009 Title

Search Results from the Web Real-Time Vehicle Classification Using Inductive Loop Signature Data https://www.researchgate.net/publication/238196961_RealTime_Vehicle_Classification_Using_Inductive_Loop_Signature_Data

1

Selected Search Results from the ASCE Civil Engineering Database Tests of Traffic Sensors and Telemetry Services Read More: http://ascelibrary.org/doi/abs/10.1061/40799%28213%2993

Selected Search Results from Transportation Research Information System (TRIS) Database Search Strategy: # Searches

Results Annotations

1 loop detector*.mp. [mp=abstract, title, heading word, accession number]

1579

2 loop sensor*.mp. [mp=abstract, title, heading word, accession number]

76

3 1 or 2 1620 4 loop signature technology.mp. [mp=abstract, title, heading word, accession number] 1 5 vehicle classification.mp. [mp=abstract, title, heading word, accession number]

977

6 3 and 5

96

7 single loop.mp. [mp=abstract, title, heading word, accession number] 8 6 and 7

115 25

Result 1. Title Author Source Publisher URL

Abstract

Vehicle Classification Using the Discrete Fourier Transform with Traffic Inductive Sensors. Lamas Seco Jose J.; Castro Paula M.; Dapena Adriana; Vazquez Araujo Francisco J Sensors. 2015. 15(10) pp 27201-27214 (Figs., Phots., Tabs.) MDPI AG http://dx.doi.org/10.3390/s151027201 Inductive Loop Detectors (ILDs) are the most commonly used sensors in traffic management systems. This paper shows that some spectral features extracted from the Fourier Transform (FT) of inductive signatures do not depend on the vehicle speed. Such a property is used to propose a novel method for vehicle classification based on only one signature acquired from a sensor single-loop, in contrast to standard methods using two sensor loops. The authors proposal will be evaluated by means of real inductive signatures captured with the authors hardware prototype.

Publication 2015 Year

2

Result 2. Title Author Source Publisher URL

Abstract

Mutually coupled multiple inductive loop system suitable for heterogeneous traffic. Ali S Sheik Mohammed; George Boby; Vanajakshi Lelitha IET Intelligent Transport Systems. 2014/8. 8(5) pp 470-478 (Figs., Phots., Refs., Tabs.) Institution of Engineering and Technology http://dx.doi.org/10.1049/iet-its.2013.0055 A magnetically coupled multiple inductive loop detector system is presented in this study. Automated detection, classification and measurement of speed of vehicles are challenging tasks, in a no-lane disciplined and heterogeneous traffic. This study proposes an inductive loop sensor wherein multiple numbers of small loops are placed within a large outer loop, for measurement of traffic parameters under such traffic. In the new system the outer loop alone is connected to the measurement unit and all the small loops are coupled inductively to the outer loop. This scheme is simple and effective and can be employed to convert an existing single loop system to a multiple loop system, suitable for heterogeneous traffic. The measurement is based on a synchronous detection method. A special excitation that ensures parallel resonance of the inductive system is employed, which keeps the power consumption minimum. The new system correctly sensed the vehicles, categorised and counted them in undisciplined traffic. The proposed system has been also extended to detect the direction of travel and speed of the vehicles. Results from the prototype developed were found to be accurate proving its practicality in real time traffic monitoring and intelligent transportation system (ITS) applications under a heterogeneous scenario.

Publication 2014 Year Result 3. Title Author Source Publisher URL

Abstract

Vehicle-Classification Algorithm Based on Component Analysis for Single-Loop Inductive Detector. Meta S; Cinsdikici M G IEEE Transactions on Vehicular Technology. 2010/7. 59(6) pp 2795-2805 (Refs.) Institute of Electrical and Electronics Engineers (IEEE) http://dx.doi.org/10.1109/TVT.2010.2049756 This paper presents a novel vehicle-classification algorithm that uses the time-variable signal generated by a single inductive loop detector. In earlier studies, the noisy raw signal was fed into the algorithm by reducing its size with rough sampling. However, this approach loses the original signal form and cannot be the best exemplar vector. The developed algorithm suggests three contributions to cope with these problems. The first contribution is to clear the noise with 3

discrete Fourier transform (DFT). The second contribution is to transfer the noiseless pattern into the Principal Component Analysis (PCA) domain. PCA is exploited not only for decorrelation but for explicit dimensionality reduction as well. This goal cannot be achieved by simple raw data sampling. The last contribution is to expand the principal components with a local maximum (L sub max ) parameter. It strengthens the classification accuracy by emphasizing the undercarriage height variation of the vehicle. These parameters are fed into the three-layered backpropagation neural network (BPNN). BPNN classifies the vehicles into five groups, and the recognition rate is 94.21%. This recognition rate has performed best, compared with the methods presented in published works. Publication 2010 Year Result 4. Title Author Source Publisher URL

Abstract

Length-based vehicle classification using event-based loop detector data. Liu Henry X; Sun Jie Transportation Research Part C: Emerging Technologies. 2014/1. 38(0) pp 156-166 (Figs., Refs., Tabs.) Elsevier http://dx.doi.org//10.1016/j.trc.2013.11.010 Length-based vehicle classification is an important topic in traffic engineering, because estimation of traffic speed from single loop detectors usually requires the knowledge of vehicle length. In this paper, the authors present an algorithm that can classify vehicles passing by a loop detector into two categories: long vehicles and regular cars. The proposed algorithm takes advantage of event-based loop detector data that contains every vehicle detector actuation and de-actuation "event", therefore time gaps between consecutive vehicles and detector occupation time for each vehicle can be easily derived. The proposed algorithm is based on an intuitive observation that, for a vehicle platoon, longer vehicles in the platoon will have relatively longer detector occupation time. Therefore, longer vehicles can be identified by examining the changes of occupation time in a vehicle platoon. The method was tested using the event-based data collected from Trunk Highway 55 in Minnesota, which is a high speed arterial corridor controlled by semi-actuated coordinated traffic signals. The result shows that the proposed method can correctly classify most of the vehicles passing by a single loop detector.

Publication 2014 Year Result 5. Title Author

Gaussian Mixture Model-Based Speed Estimation and Vehicle Classification Using SingleLoop Measurements. Lao Yunteng; Zhang Guohui; Corey Jonathan; Wang Yinhai

4

Source Publisher URL

Abstract

Journal of Intelligent Transportation Systems. 2011/10. 16(4) pp 184-196 (Figs., Refs., Tabs.) Taylor & Francis http://dx.doi.org/10.1080/15472450.2012.706196 Traffic speed and length-based vehicle classification data are critical inputs for traffic operations, pavement design and maintenance, and transportation planning. However, they cannot be measured directly by single-loop detectors, the most widely deployed type of traffic sensor in the existing roadway infrastructure. In this study, a Gaussian mixture model (GMM)based approach is developed to estimate more accurate traffic speeds and classified vehicle volumes using single-loop outputs. The estimation procedure consists of multiple iterations of parameter correction and validation. After the GMM is established to empirically model vehicle on-times measured by single-loop detectors, the optimal solution can be initially sought to separate length-based vehicle volume data. Based on the on-time of the separated short vehicles from the GMM, an iterative process will be conducted to improve traffic speed and classified volume estimation until the estimation results become statistically stable and converge. This method is straightforward and computationally efficient. The effectiveness of the proposed approach was examined using data collected from several loop stations on Interstate 90 in the Seattle area. The traffic volume data for three vehicle classes are categorized based on the proposed method. The test results show the proposed GMM approach outperforms the previous models, including conventional constant g-factor method, sequence method, and moving median method, and produces more reliable, accurate estimates of traffic speeds and classified vehicle volumes under various traffic conditions.

Publication 2011 Year Result 6. Title Author Corporate Author Source URL

Abstract

Estimation of Multiclass and Multilane Counts from Aggregate Loop Detector Data. Yuan Yufei; Wilson R Eddie; Van Lint Hans; Hoogendoorn Serge Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA Conference Title: 91st Annual Meeting. Location: Washington.Sponsored by: Transportation Research Board.Held: 20120122-20120126. 2012. 12p (Figs., Refs., Tabs.) http://pubsindex.trb.org/orderform.html Lane utilisation on the highway is subtly affected by Dynamic Traffic Management (DTM) systems such as speed controls and lane management. To optimise the operation of DTM, a better understanding of lane utilisation is required: in particular, of how the flows of different vehicle classes (passenger cars, lorries etc.) vary across the carriageway. Most loop detector systems do not collect this multi-lane multi-class count data. Therefore this paper develops a procedure for estimating multi-lane multi-class counts from a variety of standard aggregate loop data formats from around the world. The estimation procedure involves the inference of multilinear regression laws that relate multi-lane multi-class data to standard aggregate formats. The regression laws themselves then need to be trained with small samples of individual vehicle data 5

on a site-by-site basis. Preliminary results show that the estimation procedure works rather well, even when the input data is minimal - the extreme case being that of (US-style) single loop data when only flow and occupancy are available on a by-lane basis. Furthermore, an error analysis indicates that very small amounts of individual vehicle data are sufficient to train the estimator, provided they contain a representative mix of the flow behaviours at the site in question. Further work is required for the practical development of the tool, but it appears to have a wide range of potential uses for both researchers and practitioners alike. Publication 2012 Year Result 7. Title Author

Optimal Loop Placement and Models for Length-based Vehicle Classification and Stop-and-Go Traffic. Wei Heng; Ai Qingyi; Eustace Deogratias; Yi Ping

University of Cincinnati, College of Engineering and Applied Science, Cincinnati, OH, 45221-0071, USA, Ohio Transportation Consortium, University of Akron, 302 Buchtel Common, Akron, OH, Corporate 44325, USA, Research and Innovative Technology Administration, 1200 New Jersey Avenue, SE, Author Washington, DC, 20590, USA

Source

This research was sponsored by the U.S. Department of Transportation, University Transportation Centers Program.

URL

http://www.otc.uakron.edu/docs/2009%20OTC%20Project%20Final%20Report_Wei%20et%20al[1 ].pdf

Abstract

Inductive loops are widely used nationwide for traffic monitoring and as a data source for a variety of needs. The loop data have been used for vehicle length-based classification in many states including Ohio. The dual-loop detector consists of two single loop detectors which are placed apart at a fixed short distance, enabling the dual-loop detector data to be a potential real-time data source for speed and vehicle classifications. The existing dual-loop length-based vehicle classification model has been well evaluated against free flow traffic but is not suitable for nonfree flow traffic conditions (such as synchronized and stop-and-go congestion states). This project identifies the performance of the existing length-based vehicle classification models under various traffic conditions, and develops new models for congested traffic using dual-loop data. In order to ensure the right use of the new models under different traffic conditions, correct identification of varied traffic flow states is a critical need. For this purpose, an algorithm for identifying three traffic states, namely, free flow, synchronized flow, and stop-and-go flow, has been developed. A heuristic approach is employed for developing this algorithm with combination of occupancy and speed which directly result from the dual-loop data. Thresholds of variables involved in the algorithm are recommended based on the statistical analysis of the data gained from the sampling dual-loop stations in I-71/I70 in Columbus, Ohio. In addition, loop standards of layout and installation method have been collected from 17 states in the United States. Brief analysis of the collected standards is conducted to provide fundamental information for future evaluation.

Publicatio 2011 6

n Year Result 8. Title Author Corporate Author Source URL

Abstract

Estimating Truck Traffic Speed from Single-Loop Detector Data. Zhu Weihua; Boriboonsomsin Kanok; Barth Matthew J Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA Conference Title: 89th Annual Meeting. Location: Washington.Sponsored by: Transportation Research Board.Held: 20100110-20100114. 2010. 16p (5 Figs., 19 Refs., 4 Tabs.) http://pubsindex.trb.org/orderform.html Estimating truck traffic speed is a necessary step for constructing greenhouse gas emission heavy-duty trucks inventories. Embedded inductive loop detectors are the most prevalent infrastructure component for traffic monitoring in the U.S. In this paper, an algorithm for real time estimation of truck traffic speed from single-loop detector data is proposed. Conventional estimation algorithms (e.g. traffic speed estimation, truck classification, and truck traffic volume estimation) typically have their poorest performance during congestion compared to free flow conditions. Therefore, in this paper, we only focus on congested traffic by using a template matching algorithm to address the fluctuation of individual vehicle speeds during congestion. Experiments are done using freeway traffic data from the Berkeley Highway Laboratory, and the results verify the effectiveness of the proposed algorithm. An average absolute error of 3.5 mph to 5 mph during congestion of the truck traffic speed estimation is observed. The proposed template matching algorithm also has a potential application for vehicle classification.

Publication 2010 Year Result 9. Title Author

Length Based Vehicle Classification on Freeways from Single Loop Detectors. Coifman Benjamin

Nextrans, Region V Regional University Transportation Center, West Lafayette, IN, 47906, USA, Corporate Research and Innovative Technology Administration, Office of University Programs, 1200 New Author Jersey Avenue, SE, Washington, DC, 20590

Source URL

Abstract

This document is disseminated under the sponsorship of the Department of Transportation, University Transportation Centers Program. http://www.purdue.edu/dp/nextrans/docs/completedprojects/Final%20Report%20003.pdf Roadway usage, particularly by large vehicles, is one of the fundamental factors determining the lifespan of highway infrastructure, e.g., as evidenced by the federally mandated Highway Performance Monitoring System (HPMS). But the complexity of Weigh in Motion (WIM) and other classification stations makes them difficult and costly to maintain. Some of the classification 7

stations employ axle counters, but the least expensive of these stations use dual loop detectors to measure vehicle length and classify vehicles based on this measurement. To date, collecting reliable length data from single loop detectors has been considered impossible due to the noisy speed estimates. Single loop detectors promise to be an inexpensive alternative to spread classification coverage to the existing count stations and existing traffic operations detector stations. By extending classification to the relatively high density of real time traffic monitoring stations in urban areas, the classification work could allow these urban traffic management systems to better monitor freight traffic within the metropolitan areas. The research seeks to develop a means to reliably classify vehicles using estimated vehicle length from single loop detectors. Single loop detectors are the most common vehicle detector, yet they are not used for vehicle classification due to the inherent noise in the individual vehicle length estimates. This work has developed a means to extract more reliable vehicle speed estimates from single loop detectors, and thus, vehicle length estimates as well. This new, reliable, single loop detector methodology for classifying vehicles based on estimated vehicle length is significant because it will provide a low cost means of collecting vehicle classification data by extending the capabilities of existing single loop detectors. There are thousands of single loop detectors on the freeways within the region served by NEXTRANS. Publication 2009 Year Result 10. Title Author Corporate Author Source URL

Abstract

Speed Estimation and Length Based Vehicle Classification from Freeway Single Loop Detectors. Coifman Benjamin; Kim Seoungbum Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA Conference Title: 87th Annual Meeting. Location: Washington.Sponsored by: Transportation Research Board.Held: 20080113-20080117. 2008. 26p (Figs., Phots., Refs., 1 Tabs.) http://pubsindex.trb.org/paperorderform.pdf Roadway usage, particularly by large vehicles, is one of the fundamental factors determining the lifespan of highway infrastructure. Each state typically has several dozen Weigh in Motion stations to monitor large vehicle usage. These stations are expensive to install and maintain, so they are usually supplemented with many more vehicle classification stations. Some of the classification stations employ axle counters, but the simplest of these stations use dual-loop detectors to measure vehicle length from the product of measured speed and detector on-time, and classify vehicles based on this measurement. Meanwhile, single-loop detectors are the most common vehicle detector in use to monitor traffic, both for real-time operations and for collecting census data such as Annual Average Daily Travel (AADT). New, out-of-pavement detectors seek to replace loop detectors using wayside mounted sensors, but most of these detectors emulate the operation of single-loop detectors. In either case, collecting reliable length data from these detectors has been considered impossible due to the noisy speed estimates provided by conventional data aggregation at single-loop detectors. This research refines unconventional techniques for estimating speed at a single-loop detector, yielding estimates that approach the accuracy of a dual-loop detector's measurements. Employing these speed estimation advances, this research brings length based vehicle classification to 8

single-loop detectors, (and by extension, many of the emerging out-of-pavement detectors). The research promises to extend vehicle classification to single-loop detector count stations and the many single-loop detector stations already deployed for real-time traffic management. The work also offers a viable treatment in the event that one of the loops in a dual-loop detector classification station fails. The classification methodology is evaluated against concurrent measurements from video and dual-loop detectors. To capture higher truck volumes than empirically observed, a process of generating synthetic on-times is developed. Publication 2008 Year Result 11. Title Author Corporate Author Source URL

Abstract

Real-Time Vehicle Classification Using Inductive Loop Signature Data. Jeng Shin Ting (Cindy); Ritchie Stephen G Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA Conference Title: 87th Annual Meeting. Location: Washington.Sponsored by: Transportation Research Board.Held: 20080113-20080117. 2008. 38p (4 Figs., 16 Refs., 11 Tabs.) http://pubsindex.trb.org/paperorderform.pdf Vehicle class is an important characteristic of traffic measurement, and classification information can contribute to many important applications in various transportation fields. For instance, vehicle classification is helpful to monitor heavy vehicle traffic for road maintenance and safety, to model traffic flow, and to obtain performance measurements based on each vehicle class for traffic surveillance. In this research, a real-time vehicle classification model is introduced. A heuristic method combined with decision tree and K-means clustering approaches is proposed to develop the vehicle classification model. The features used in the proposed model were extracted from Piecewise Slope Rate (PSR) values, which were obtained from single loop inductive signature data. Three vehicle classification schemes, FHWA, FHWA-I, and RTPMS (Real-time Traffic Performance Measurement System), and a dataset obtained from square single loop detectors was utilized for model development. A dataset obtained from round single loop detectors was applied to test the transferability of the proposed model. The results demonstrate that the proposed real-time vehicle classification model is not only capable of categorizing vehicle types based on the FHWA scheme, but also capable of grouping vehicles into more detailed classes. The classification model can successfully classify vehicles into 15 classes using single loop detector data without any explicit axle information. In addition, the advantages of the proposed vehicle classification model are its simplicity, employing the current detection infrastructure, and enhancing the use of single loop detectors for vehicle classification. The initial results also suggest the potential for transferability of the vehicle classification approach, and are very encouraging.

Publication 2008 Year Result 12. Title

Vehicle Classification from Single Loop Detectors. 9

Author

Corporate Author

Source URL

Abstract

Coifman Benjamin Ohio State University, Columbus, Department of Civil and Environmental Engineering and Geodetic Science, Columbus, OH, 43210, USA, Midwest Regional University Transportation Center, University of Wisconsin, 1415 Engineering Drive, Madison, WI, 53706, USA, Wisconsin Department of Transportation, 4802 Sheboygan Avenue, P.O. Box 7910, Madison, WI, 537077910, USA, Research and Special Programs Administration, 1200 New Jersey Avenue, SE, Washington, DC, 20590, USA, University of Wisconsin, Madison, Department of Civil and Environmental Engineering, 1415 Engineering Drive, Madison, WI, 53706, USA This research was funded by the U.S. Department of Transportation, University Transportation Centers Program. http://www.mrutc.org/research/0502/index.htm Vehicle classification data are important inputs for pavement maintenance, traffic modeling, and emission evaluation. Various technologies including weigh-in-motion (WIM), axle counting with piezo-electric sensors or length measurement from dual loop detectors have been used for vehicle classification. This research extends length based vehicle classification to single loop detectors. It promises a lower cost alternative as well as the potential to use existing detectors already deployed for freeway management. Of course the single loop based estimates could also be easily incorporated in a more sophisticated classification station as an independent validation of its measurements. The main challenge with single loop detector based length based classification comes from accurately estimating speed and thus, length. This study develops a methodology to make such accurate speed and length estimates ant then uses the latter to classify vehicles based on length. Performance is validated against two sources of independent ground truth data with results that approach the accuracy of dual loop detectors. In the process of generating ground truth data a few previously unknown, sight specific problems with existing vehicle classification and detection stations were found and diagnosed, e.g., pulse break-up, as discussed in the report.

Publication 2007 Year Result 13. Title Author Source Publisher URL

Video-Based Vehicle Detection and Classification System for Real-Time Traffic Data Collection Using Uncalibrated Video Cameras. Zhang Guohui; Avery Ryan Patrick; Wang Yinhai Transportation Research Record: Journal of the Transportation Research Board. 2007. (1993) pp 138-147 (Figs., Phots., 29 Refs., 1 Tabs.) Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA http://dx.doi.org/10.3141/1993-19

10

Abstract

Length-based vehicle classification data are important inputs for traffic operation, pavement design, and transportation planning. However, such data are not directly measurable by singleloop detectors, the most widely deployed type of traffic sensor in the existing roadway infrastructure. In this study a video-based vehicle detection and classification (VVDC) system was developed for truck data collection using wide-ranging available surveillance cameras. Several computer vision-based algorithms were developed or applied to extract background image from a video sequence, detect presence of vehicles, identify and remove shadows, and calculate pixel-based vehicle lengths for classification. Care was taken to handle robustly negative effects resulting from vehicle occlusions in the horizontal direction and slight camera vibrations. The pixel-represented lengths were exploited to distinguish long vehicles from short vehicles; hence the need for complicated camera calibration can be eliminated. These algorithms were implemented in the prototype VVDC system using Microsoft Visual C#. As a plug-and-play system, the VVDC system is capable of processing both digitized image streams and live video signals in real time. The system was tested at three test locations under different traffic and environmental conditions. The accuracy for vehicle detection was above 97%, and the total truck count error was lower than 9% for all three tests. This indicates that the video image processing method developed for vehicle detection and classification in this study is indeed a viable alternative for truck data collection.

Publication 2007 Year Result 14. Title Author Source Publisher URL

Abstract

Artificial Neural Network Method for Length-Based Vehicle Classification Using Single-Loop Outputs. Zhang Guohui; Wang Yinhai; Wei Heng Transportation Research Record: Journal of the Transportation Research Board. 2006. (1945) pp 100-108 (5 Figs., 24 Refs., 4 Tabs.) Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA http://dx.doi.org/10.3141/1945-12 Classified vehicle volumes are important inputs for traffic operation, pavement design, and transportation planning. However, such data are not available from single-loop detectors, the most widely deployed type of traffic sensor in the existing roadway infrastructure. Several attempts have been made to extract classified vehicle volume data from single-loop measurements in recent years. These studies used estimated speed for length calculation and classified vehicles into bins based on the calculated vehicle lengths. However, because of the stochastic features of traffic flow, deterministic mathematical equations based on certain assumptions for speed calculation typically do not work well for all situations and may result in significant speed estimation errors under certain traffic conditions. Such errors accumulate when estimated speeds are used in vehicle-length calculations and degrade the accuracy of vehicle classification. To solve this problem, an artificial neural network method was developed to estimate classified vehicle volume data directly from single-loop measurements. The proposed neural network is three-layered with a back-propagation structure. This method was tested with data collected from several loop stations on I-5 over a long duration. The proposed artificial neural network model produced reliable estimates of volumes of classified vehicles under 11

various traffic conditions. Publication 2006 Year Result 15. Title Author Corporate Author Source URL

Abstract

Robust Algorithm for Improved Dual-Loop Detection on Freeways. Zhang Xiaoping; Wang Yinhai; Nihan Nancy Lucille Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA Conference Title: 85th Annual Meeting. Location: Washington.Sponsored by: Transportation Research Board.Held: 20060122-20060126. 2006. 24p (5 Figs., 5 Refs., 4 Tabs.) http://pubsindex.trb.org/orderform.html The Washington State Department of Transportation (WSDOT) has a dual-loop detection system on its Greater Seattle freeway network to provide real-time speed and vehicle classification data in a four-bin format. The dual-loop's capability of classifying vehicles makes the dual-loop detection system a potential real-time truck data source for freight movement study. However, a previous study found that the WSDOT dual-loop detectors were consistently underreporting truck volumes, whereas the single-loop detectors were consistently over counting vehicle volumes. The effectiveness of the current dual-loop algorithm was questioned. In light of advances in controller computing capability, an improved dual-loop algorithm that can tolerate erroneous raw loop actuation signals was developed in this research. The improved dual-loop algorithm includes a noise filter and a postprocessor to screen out noise, a matching scheme to pair up on-time pulses, and an improved speed and length calculation method to increase the reliability of dual-loop outputs. When calculating speed and vehicle length, various checks are applied to test the validity of the data. If any of the checks fails, an appropriate error is flagged, but the individual vehicle data are not discarded from the total count. The data analysis conducted in this research verified the effectiveness of the noise filter and the postprocessor, the on-time pulse matching scheme, and the improved dual-loop algorithm.

Publication 2006 Year Result 16. Title Author Source Publisher Abstract

Vehicle-Classification Algorithm for Single-Loop Detectors Using Neural Networks. Ki Yong Kul; Baik Doo Kwon IEEE Transactions on Vehicular Technology. 2006/11. 55(6) pp 1704-1711 (9 Figs., 20 Refs., 4 Tabs.) Institute of Electrical and Electronics Engineers This paper presents a new algorithm for inductive loop detectors (ILD) used in vehicle 12

classification. In order to improve accuracy, the new algorithm uses back propagation neural networks. The variation rate of frequency and frequency waveform are the inputs to the neural networks in the algorithm. Five classified vehicles serve as the output. Tests assessing the new algorithm revealed a 91.5% recognition rate. When compared to the conventional method based on ILD, this new algorithm was shown to improve vehicle classification accuracy. Publication 2006 Year Result 17. Title Source Publisher URL

Abstract

Traffic and Urban Data. Transportation Research Record: Journal of the Transportation Research Board. 2006. (1945) 125p (Figs., Refs., Tabs.) Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA http://pubsindex.trb.org/images/covers/TRR1945.png This Transportation Research Record contains 13 papers on the subject of traffic and urban data. Among the specific topics discussed in these papers are the following: spacing of traffic sensors; travel time variability calculated from loop detector data; automatic vehicle identification data for studying travel time variability; cellular-based data extracting; holiday effects on traffic volumes; NCHRP 1-37A pavement design sensitivity to traffic input; traffic monitoring devices accuracy analysis; truck traffic characterization for mechanistic-empirical design; dual-loop sensitivity problems identification and correction; detection of invalid loop detector data; identification of lane-mapping errors at freeway detector stations; using artificial neural networks for length-based vehicle classification from single-loop outputs; and quartz piezoelectric weigh-in-motion sensors.

Publication 2006 Year Result 18. Title Author Source Publisher URL Abstract

Improved Dual-Loop Detection System for Collecting Real-Time Truck Data. Zhang Xiaoping; Nihan Nancy L; Wang Yinhai Transportation Research Record: Journal of the Transportation Research Board. 2005. (1917) pp 108-115 (3 Figs., 14 Refs., 3 Tabs.) Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA http://dx.doi.org/10.3141/1917-13 The Washington State Department of Transportation (WSDOT) has a loop detection system on its Greater Seattle freeway network to provide real-time traffic data. The dual-loop detectors installed in the system are used to measure vehicle lengths and then classify each detected 13

vehicle into one of four categories according to its length. The dual loop's capability of measuring vehicle length makes the loop detection system a potential real-time truck data source for freight movement studies because truck volume estimates by basic length category can be developed from the vehicle length measurements produced by the dual-loop detectors. However, a previous study found that the dual-loop detectors were consistently underreporting truck volumes, whereas the single-loop detectors were consistently overcounting vehicle volumes. As an extension of the previous study, the research project described here investigated possible causes of loop errors under nonforced-flow traffic conditions. A new dual-loop algorithm that can address these error causes and therefore tolerate erroneous loop actuation signals was developed to improve the performance of the WSDOT loop detection system. A quick remedy method was also recommended to address the dual-loop undercount problem without replacing any part of the existing system hardware or software. In addition, a laptop-based detector event data collection system (DEDAC) that can collect loop detector event data without interrupting a loop detection system's normal operation was developed in this research. The DEDAC system enables various kinds of transportation research and applications that could not otherwise be possible. Publication 2005 Year Result 19. Title Author Corporate Author Source Publisher

Abstract

DYNAMIC ESTIMATION OF FREEWAY LARGE-TRUCK VOLUMES BASED ON SINGLELOOP MEASUREMENTS. Wang Y; Nihan N L Taylor & Francis, 325 Chestnut Street, Suite 800, Philadelphia, PA, 19106, USA Journal of Intelligent Transportation Systems. 2004/7. 8(3) p. 133-141 (5 Figs., Refs., 4 Tabs.) Taylor & Francis, 325 Chestnut Street, Suite 800, Philadelphia, PA, 19106, USA Collection of large truck (LT) volume data is very important for accurate analysis of traffic stream characteristics in transportation planning and engineering. However, since LT travel patterns are seasonal, data obtained by surveys conducted for a short time period every one to three years may not be adequate. Therefore, the ability to collect such data continuously via loop detectors, especially single loop detectors, would be useful. This paper presents an algorithm for estimating LT volumes using only single-loop outputs. LT volumes estimated by the proposed algorithm were compared with those observed by dual-loop detectors. Results showed that the two LT volume series fit each other very well, especially when traffic volume was low. Possible estimation errors in using the algorithm are discussed.

Publication 2004 Year Result 20. Title Author

CAN SINGLE-LOOP DETECTORS DO THE WORK OF DUAL-LOOP DETECTORS?. Wang Y; Nihan N L 14

Corporate Author Source Publisher

Abstract

Publication Year

American Society of Civil Engineers, 1801 Alexander Bell Drive, Reston, VA, 20191-4400, USA Journal of Transportation Engineering. 2003/3. 129(2) p. 169-176 (8 Figs., Refs., 4 Tabs.) American Society of Civil Engineers, 1801 Alexander Bell Drive, Reston, VA, 20191-4400, USA Real-time speed and vehicle-classification data are important inputs for modern freeway traffic control and management systems. However, these data are not directly measurable by singleloop detectors. Although dual-loop detectors provide speeds and classified vehicle volumes, there are too few of them on our current freeway systems to meet the practical ATMS (Advanced Traffic Management System) needs, and the cost of upgrading from a single-loop detector to a dual-loop detector is high. This makes it extremely desirable to develop appropriate algorithms to make single-loop detectors capable of performing the tasks of double loops. This paper presents just such an algorithm, i.e., one that uses single-loop measurements to provide accurate speed and vehicle-classification estimates. There are 3 steps in the algorithm: 1) to separate intervals with long vehicles (LVs) from those without; 2) to use measurements of intervals without LVs for speed estimation; and 3) to identify LV volumes for the intervals with LVs using the estimated speed. Preliminary tests for both spatial transferability and temporal transferability of the algorithm were performed, and the results were encouraging. 2003

Result 21. Title Author

Corporate Author

Source Publisher URL

Abstract

FIELD INVESTIGATION OF ADVANCED VEHICLE REIDENTIFICATION TECHNIQUES AND DETECTOR TECHNOLOGIES. PHASE 1. Ritchie Stephen Graham; Park Seri; Oh Cheol; Sun Carlos Partners for Advanced Transit and Highways (PATH), University of California, Richmond, CA, 94804-4648, USA, Partners for Advanced Transit and Highways (PATH), University of California, 1357 South 46th Street, Building 452, Richmond, CA, 94804-4648, USA, University of California, Irvine, Institute of Transportation Studies, Irvine, CA, 92697, USA, University of Missouri, Columbia, Department of Civil and Environmental Engineering, Columbia, MO, 65211-2200, USA, California Department of Transportation, 1120 N Street, P.O. Box 942673, Sacramento, CA, 95814, USA PATH research report ; UCB-ITS-PRR-2002-15. 2002. xii, 87 p. Partners for Advanced Transit and Highways (PATH) http://www.path.berkeley.edu/PATH/Publications/PDF/PRR/2002/PRR-2002-15.pdf This report describes a study in which a real-time traffic surveillance system based on vehicle reidentification technology utilizing vehicle inductive signatures was implemented. Improved estimates of fundamental real-time traffic parameters such as speed volume, and vehicle class from single loop detector inductive signatures were derived. The report discusses how an 15

inductive loop detector (ILD)-based vehicle reidentification system was expanded at a major signalized intersection in Irvine, California, to address reidentification of turning vehicles in addition to through vehicles. New techniques were developed for on-line real-time intersection level of service estimation. The report presents an overview of a capability that was developed for communicating real-time traffic performance data to operators at a Transportation Management Center (TMC). It describes the development of a prototype-real-time web site for Internet-based access to performance data from the study intersection. Results from the initial testing of a new state-of-the art detector card are also reported. Publication 2002 Year Result 22. Title Author Corporate Author Source Publisher URL

Abstract

REAL-TIME TRAFFIC MEASUREMENT FROM SINGLE LOOP INDUCTIVE SIGNATURES. OH S; Ritchie S G; Oh C Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA Transportation Research Record. 2002. (1804) p. 98-106 (8 Figs., 14 Refs., 8 Tabs.) Transportation Research Board, 500 Fifth Street, NW, Washington, DC, 20001, USA http://dx.doi.org/10.3141/1804-14 Accurate traffic data acquisition is essential for effective traffic surveillance, which is the backbone of advanced transportation management and information systems (ATMIS). Inductive loop detectors (ILDs) are still widely used for traffic data collection in the United States and many other countries. Three fundamental traffic parameters--speed, volume, and occupancy-are obtainable via single or double (speed-trap) ILDs. Real-time knowledge of such traffic parameters typically is required for use in ATMIS from a single loop detector station, which is the most commonly used. However, vehicle speeds cannot be obtained directly. Hence, the ability to estimate vehicle speeds accurately from single loop detectors is of considerable interest. In addition, operating agencies report that conventional loop detectors are unable to achieve volume count accuracies of more than 90% to 95%. The improved derivation of fundamental real-time traffic parameters, such as speed, volume, occupancy, and vehicle class, from single loop detectors and inductive signatures is demonstrated.

Publication 2002 Year Result 23. Title Author Source

Monitoring a freeway network in real time using single-loop detectors. ZHANG,X; WANG,Y; NIHAN,NL; DONG,H INTERNATIONAL JOURNAL OF VEHICLE DESIGN. 2005. 1(1/2) pp119-130

16

Publisher

Abstract

INDERSCIENCE ENTERPRISES LTD, WORLD TRADE CENTER BUILDING, 29 ROUTE DE PRE BOIS, CASE POSTALE 896, GENEVE, SUISSE With the widespread adoption of automated traffic sensors, the increase of computer processing power, and the evolution and growth of the global Internet, our ability to monitor traffic and convey traffic information in real-time has been dramatically advanced. To the authors' knowledge, most of the current traffic information systems mainly provide general traffic flow information. Vehicle classification data, such as truck volumes, that should be key inputs for good transportation planning, freight mobility analysis, roadway geometric and structural design and traffic control and operation, are often excluded or only roughly estimated. To address this problem, this paper describes the design and implementation of a traffic monitoring and information system recently developed at the University of Washington that visually conveys speed and vehicle classification information, obtained by processing real-time single-loop measurements, to the general public through the client/server computer architecture. The development of this system makes real-time monitoring of truck volume data on a freeway network possible.(A)

Publication 2005 Year Result 24. Title Author

VEHICLE CLASSIFICATION USING DIGITAL INDUCTIVE LOOP VEHICLE DETECTORS. PYE,KJ

Source

12TH ARRB CONFERENCE, HOBART, TASMANIA, 27-31 AUGUST 1984. PROCEEDINGS. 1984. 12(4) pp208-16 (15 Figs., 6 Refs.)

Publisher

AUSTRALIAN ROAD RESEARCH BOARD, 500 BURWOOD HIGHWAY, VERMONT SOUTH, VICTORIA, 3133, AUSTRALIA

Abstract

MODERN INDUCTIVE LOOP VEHICLE DETECTORS USE MICROPROCESSORS AND SAMPLING OF LOOP SIGNALS TO IMPROVE THE PERFORMANCE OF THE DETECTOR, BUT STILL DISCARD MUCH OF THE AVAILABLE INFORMATION. THIS PAPER DISCUSSES SOME OF THE POSSIBLE USES OF THE INFORMATION AVAILABLE FROM CERTAIN LOOP GEOMETRIES IN RELATION TO VEHICLE CLASSIFICATION AND IDENTIFICATION. COMPARISONS ARE MADE BETWEEN A VEHICLE CLASSIFICATION SYSTEM BASED ON A DIGITAL INDUCTIVE LOOP DETECTOR USING A SINGLE LOOP WITH A SPECIALLY CHOSEN GEOMETRY AND EXISTING COMMERCIAL SYSTEMS WHICH TYPICALLY USE TWO INDUCTIVE LOOPS AS WELL AS AN AXLE DETECTOR. SOME TYPICAL INDUCTANCE- TIME PROFILES ARE DISCUSSED, AND THE PROFILES ARE SHOWN TO BE REPEATABLE BETWEEN DIFFERENT RUNS OF THE SAME VEHICLE. THE CORRESPONDENCE BETWEEN PROFILE AND VEHICLE SHAPE IS INVESTIGATED, TOGETHER WITH AN EXAMINATION OF THE DIFFERENCES BETWEEN PROFILE SHAPES OF DIFFERENT TYPES OF VEHICLE. THE LIMITATIONS IN THE INFORMATION AVAILABLE FROM THE PROFILES ARE NOTED AND SUGGESTIONS MADE AS TO POSSIBLE FUTURE INVESTIGATIONS (A). THE NUMBER OF THE COVERING ABSTRACT FOR THE CONFERENCE IS IRRD NO 277326.

17

Publication 1984 Year Result 25. Title Author Source Publisher

Abstract

LOOP SENSORS FOR VEHICLE CLASSIFICATION. DAVIES,P; SALTER,DR; BETTISON,M TRAFFIC ENG CONTROL. 1982/02. 23(2) pp55-9 (9 Figs., 4 Refs.) LONDON, UNITED KINGDOM LABORATORY AND FIELD TESTS HAVE BEEN CARRIED OUT TO IMPROVE LOOP SENSOR PERFORMANCE IN VEHICLE CLASSIFICATION UNDER FREE-FLOW TRAFFIC CONDITIONS. THE LOOPS ARE USED WITH THE TRRL MICROPROCESSOR SYSTEM TO CLASSIFY VEHICLES ACCORDING TO LENGTH WHEELBASE AND CHASSIS HEIGHT. PROBLEMS OCCUR IN DISTINGUISHING BETWEEN CARS AND CERTAIN GOODS VEHICLES WHEN LANE DISCIPLINE IS POOR. AN OUTLINE IS GIVEN OF THE OPERATING PRINCIPLES OF LOOP DETECTION SHOWING HOW THE COMPONENTS OF THE MAGNETIC FIELD ARE UTILISED TO DETECT LATERAL AND VERTICAL CHARACTERISTICS OF THE PASSING VEHICLES. THE MERITS OF SEVERAL DIFFERENT LOOP CONFIGURATIONS INCLUDING HEXAGONAL LOOPS, DOUBLE TRIANGULAR LOOPS AND DOUBLE AND SINGLE RECTANGULAR LOOPS ARE DISCUSSED. LABORATORY EXPERIMENTS SHOWED THAT IT WAS NOT POSSIBLE TO PRODUCE AN IDEAL SINGLE LOOP AND THAT THE RELIABLE DETECTION OF BICYCLES WAS NOT SIMPLE. LOOP SELECTION DEPENDS UPON THE MOST DESIRABLE CHARACTERISTICS REQUIRED AND IT WAS THOUGHT THAT THE DOUBLE RECTANGULAR LOOP, WITH A 200MM SEPARATION BETWEEN THE CENTRAL CONDUCTORS, WAS THE BEST FOR DETECTING HIGH SIDED ARTICULATED VEHICLES WITH GOOD RESOLUTION OF THE CHASSIS HEIGHTS OF CARS, VANS AND MEDIUM GOODS VEHICLES.

Publication 1982 Year

18