I. Problem Statement Number

I. PROBLEM STATEMENT NUMBER

(To be completed by NCHRP)

II. PROBLEM TITLE

Using Automated Detectors to Maximize Safety for Pedestrians and Bicyclists at Signalized Intersections

III. RESEARCH PROBLEM STATEMENT

At a signalized intersection, pedestrians and bicyclistsmust be safely accommodated by signal phasing and timing mechanism. After the “WALK” interval, the “Flashing DONT WALK (FDW)”interval is released for a predetermined duration, which is followed by a “Steady DONT WALK” interval to stop pedestrians. The Highway Capacity Manual (HCM) prescribes that the parallel vehicular green must be at least equivalent to “WALK” plus “FDW” for pedestrian safety sake. Hence, the design walking speed is critical to determine the amount of crossing time actually provided to pedestrians and bicyclists. However, the MUTCD-required 3.5 ft/sec (or lower speed) forthe FDW timing may be very inefficient for the phases when there are no slow pedestrians or bikes to use the longer FDW display and the unused FDW time would have been efficiently used for vehicles to reduce queuing time.

Automated pedestrian detection (APD) is increasingly applied in Europe and Australia with Puffin (Pedestrian User-Friendly Intelligent) crossing systemsto activate pedestrian phases, prolong crossing clearance time for slower pedestrians and bicyclists,and retract pedestrian phases when not needed. A British study show that “Puffin crossings have a beneficial effect on safety giving a mean reduction in pedestrian accidents of 27% and a reduction of 19% overall” (UK DFT, 2009). However, ithas not been widely accepted in North America yet. Therefore, it is necessary to harness APD technology to maximize pedestrian and bike safety in an intelligent way.

Effective use of this promising technology relies on: (1) Appraisal of its efficacy, reliability, and robustness.There is inadequate understanding of APD technologies in the repertoire which spans infrared/heat sensors, microwave, pressure mats, and computer-assisted video. A multitude of field tests have produced inconsistent results which are interpreted and synthesized in limited ways.(2) Integration of pedestrian detectors into current traffic signal systems tointelligently balance the multimodal needs.Traffic signal control methods should be adaptable to the complexity of using APD data inputs. (3) Satisfyinga wide spectrum of technical concerns in practical implementation.For example, there are questions fromlocal traffic management personnel about the reliability, maintenance requirements, liability exposure, and the accessibility requirements of such APD devices.

IV. LITERATURE SEARCH SUMMARY

Many research projects have been conducted to address pedestrian safety related issues at crosswalks, but very few studies specifically address both the safe FDW timing and vehicle efficiency issues stated above. Furthermore, the literature review reinforces the promising value of APD and its seamless blend with currenttraffic signal systems. For example, a 2008 FHWA (Federal Highway Administration) Pedestrian Safety Report to Congressstressed the potential of automated (or passive) pedestrian detection to ameliorate safety. However, it is also found that these technologies “require additional research and extensive field testing to demonstrate and evaluate the benefits of deploying the systems.” It pointed to concerns about costs and reliability, as well as the gap between limited U.S. experience and wider European and Australian popularity. A Pedestrian and Bicyclist Safety and Mobility International Scan Team (2009) listed passive detection of pedestrians at signalized crossings as one of several “innovative traffic signal features and design practices that have the potential to improve pedestrian safety in the U.S.”.North American studies on APD technologies are inadequate but gradually increasing. For instance, a recent field assessment of varied sensors in College Station, TX, found that infrared and microwave sensors had error rates ranging from 9% to 39% (Turner et al., 2007). Researchers concluded that “the accuracy of the sensors appeared to be very location-specific, in that pedestrian detection can be more effective in certain situations in which the pedestrian travel area is constrained.” San Francisco’s PedSafe project tested a range of pedestrian safety measures, including using the Econolite Autoscope™ video system to adjust crossing times and comparing flashing beacon installations: one actuated by push buttons, the other by infrared bollards (SFMTA, 2008).More major progresses with testing stereo infrared systems have been made by Migma Systems with funds from the FHWA.

V. RESEARCH OBJECTIVES

This project will expand further on recent accomplishments of a technical committee of the Institute of Transportation Engineers (ITE). The ITE committee finalized a broad state-of-the-art report on APD, considering the literature and the views of transportation engineers and planners (including a major on-line survey). This proposed research work will develop case studies and test the integration of APD into current vehicle-actuated traffic signal systems widely installed at isolated intersections.At present, the similar issue with multiple intersections under coordinated signalizations is out of the study scope.

The focus will be on dynamic detection-based adjustment of FDW timing instantaneously to reflect real-time needs and responsive adjustment of other timing parameters.The research objective is to appraise and enhance the effectiveness of APD technologies in the field and also realize aholistic incorporation ofAPD into a specific deployable traffic signal system. The research would not only quantify the real-world benefits of this multimodal detection approach but also identify the possible barriers againstprevalentdeployments of successful APD technologies.Due to the complex domain for the issue, the proposed research will be divided into multiple phases which could be implemented sequentiallyand/or simultaneously, counting on availability of resources and work progresses.

Phase 1:Effectiveness and Reliability Field Tests of APD Systems

The results of literature review have helped identify the most promising technologies, how to use them, and criteria for future tests. Particularly, the extensive European and Australian experience could be tapped out more fully for future field tests. Findings on maintenance and reliability issues should also be useful in minimizing potential drawbacks which could influence field tests.

1)Based on existing typical hardware (e.g., 2070 controller etc.) and software (e.g., D4 etc.), conduct field experiments with different APD systems, including computer-aided video detection systems (e.g., Econolite Autoscope etc.) to adjust pedestrian crossing time to accommodate slower pedestrians. The test intersections would be selected to provide a range of traffic conditions with different vehicle/pedestrian volumes and crossing distances.

2)Intensive data collection on pedestrian and driver behavior changes to answer questions such as:

a)Whether pedestrians become aware of the automated system and then change their behaviors?

b)What are impacts of providing information to pedestrians about automated detection?

c)How effectively the tested detection devices perform (false and missed activations)?

3)Comprehensive test record keeping of operational costs, maintenance, reliability, legal, and accessibility issues related to pedestrian detection systems tested.

4)Experiment with other promising pedestrian detection devices with limited application experience (e.g., heat sensors, long-wave infrared sensors, or millimeter wave detectors etc.) in similar ways as above.

Phase 2: Hardware-in-the-loop Simulation Test

1)Hardware-in-the-loop simulation (HILS) provides a unique and effective means of testing and evaluating the intelligentsignal control logic newly embedded into a real-world controller in a high-fidelity simulation environment before its field implementation.The controller interface device (CID) plays the crucial role in establishing the interface between the microsimulation software test-bed and a real-world traffic signal controller, since it is an embedded system which relays detector data from simulation to a controller and returning phase information back to simulation. Therefore, one task in this phase is to develop a new CID and its support applications which effectively satisfy not only the detection input needs for pedestrian walking speeds but also the similar needs for vehicles.

2)The subsequent task is to transplant the new signal system prototypeonto an actual signal controller to implement an offline HILS-based effectiveness test under a wide spectrum of traffic conditions characterized by different vehicle/pedestrian volumes and crossing distances.

Phase 3: Field Test

1)Establish the whole traffic management architecture in the field through integrating the pedestrian and vehicle detection equipment with the new signal systems embedded in a real-world controller.

2)Conduct a sufficient number of system performance tests at a typical field site in urban settings under various traffic and weather conditions (e.g., AM/PM peak hours, non-peak hours, rainy days, wintry days, etc.), with the intention to evaluate efficacy, safety, efficiency, reliability, and sensitivityin terms of the rates for breakdown, false call, false detection, etc.

Phase 4: Report

The final report for the proposed project will describe the phase work, results, and findings. The research information will be disseminatedatvaried media such as conference, webinar, website, TV andradio broadcasting and so forth.

VI. ESTIMATE OF PROBLEM FUNDING AND RESEARCH PERIOD

Recommended Funding: $300,000 - $500,000

Research Period: 3.0 years (36 months)

VII. URGENCY, PAYOFF POTENTIAL, AND IMPLEMENTATION

High urgency.The principal contribution of this research is the first realization of asystematic, real-time,dynamic, andadaptive accommodation for the safety and operational needs of all intersection users.

VIII. PERSON(S) DEVELOPING THE PROBLEM

TRB Committee on Pedestrians (ANF 10).

Primary Author: George Lu, Ph.D.

Research Fellow (Transportation Engineering)

The University of Michigan (U-M)

Ann Arbor, MI, USA

E-mail:

Reviewers:Paula Reeves, Washington State Department of Transportation

Dan Painter, Virginia State Department of Transportation

Matthew Ridgway, Fehr & Peers, CA

Beeze Benson, Access for Blind, MD

Frank Markowitz, San Francisco Municipal Transportation Agency, CA

Contact Information:

Ilona Kastenhofer, TRB Pedestrian Research Subcommittee Chair, Virginia Department of Transportation, Virginia Center for Transportation Innovation and Research (434) 293-1981, .

IX. PROBLEM MONITOR

(to be completed)

X. DATE AND SUBMITTED BY

To be submitted by the TRB Committee on Pedestrians (ANF 10)

References

AutoEvolution web site (2009). How Volvo’s Pedestrian Protection System Works.

Bechtel, A., Geyer, J., and Ragland, D. (2004).A Review of ITS-Based Pedestrian Injury Countermeasures.Compendium of the Annual Meeting of the Transportation Research Board, Washington, DC.

Eco-Compteur (2008).Eco-counter web site.

Federal Highway Administration (2008).Pedestrian Safety: Report to Congress. FHWA web site.

Fitzpatrick, K., S. Turner and M. Brewer.Improving Pedestrian Safety at Unsignalized Roadway Crossings.ITE Journal, Institute of Transportation Engineers, Washington, DC, May 2007, pp. 34-41.

Gibson, D., B. Lang, B. Ling, U. Venkataraman, and J. Yang. (2009). Detecting Pedestrians. Public Roads, 73 (September/October 2009).

Markowitz, F., Montufar, J., and Steindel, M. (2009).Automated Pedestrian Detection: State of the Art.Presented at the Institute of Transportation Engineers (ITE) Annual Meeting, San Antonio, TX.

Pedestrian and Bicyclist Safety and Mobility International Scan Team (2009). International Scan Summary Report on Pedestrian and Bicyclist Safety and Mobility.Sponsored by Federal Highway Administration, American Association of State Highway Transportation Officials, and National Cooperative Highway Research Program.

SFMTA (San Francisco Municipal Transportation Agency) with the University of California at Berkeley Traffic Safety Center (2008). PedSafe Phase II Implementation Report.Performed for the Federal Highway Administration.

TRL Limited (2005). Puffin Crossing Operation and Behavior Study.Transport for London Published Project Report PPR239.

Turner, S. et al. (2007).Testing and Evaluation of Pedestrian Sensors.Texas Transportation Institute, Report No.SWUTC/07/167762-1.

Walkinginfo.org (2007).Accessible Pedestrian Signals: Synthesis and Guide to Best Practice: Passive Pedestrian Detection.

Zegeer, C.V., Opiela, K.S., and Cynecki, M.J., Pedestrian Signalization Alternatives, Report No. FHWA-RD-83-102, Federal Highway Administration, Washington, DC, July 1985.

1