U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-HRT-05-080
Date: May 2006
Pedestrian Access to Roundabouts: Assessment of Motorists' Yielding to Visually Impaired Pedestrians and Potential Treatments to Improve Access
PDF Version (1.14 MB)
PDF files can be viewed with the Acrobat® Reader®
CHAPTER 1. INTRODUCTION
This report describes two related studies intended to address double-lane roundabout accessibility issues for visually impaired pedestrians. The first study was conducted on a closed course to evaluate the feasibility of a pavement treatment to alert visually impaired pedestrians when vehicles have yielded to them. The second study examined drivers' yielding behavior at a two-lane roundabout and the effectiveness of the same roadway treatment in an operational environment.
People who are visually impaired may experience difficulty using roundabout crosswalks.(1,2) Recent research has documented this difficulty, at least for cases where traffic volumes are high. (3) The problem, as summarized by the U.S. Access Board, consists of three parts: (1)
The Americans with Disabilities Act (ADA) requires equal access to transportation facilities. The U.S. Access Board, which is responsible for developing guidelines for implementing the ADA, currently is considering guidelines for providing access at roundabouts. (4) The studies reported here are intended to contribute to an empirical basis for those guidelines.
According to statistics compiled by the National Center for Health Statistics of the U.S. Centers for Disease Control and Prevention, as summarized by the American Foundation for the Blind TM, in 1996, there were approximately 1.3 million legally blind people in the United States. (5) Of these, approximately 109,000 people use long white canes, and another 7,000 use guide dogs. Approximately two-thirds of the people who use a long white cane or a guide dog are less than 65 years old, and about half are employed full time. Although statistics are not available on the number of visually impaired individuals who travel independently on streets and sidewalks, the statistics on long cane and guide dog users suggest that this number is around 100,000. As the number of roundabout intersections grows, so will the importance of accessibility. Furthermore, treatments to improve access at roundabouts might also apply to other types of uncontrolled crosswalks, such as midblock crossings and right-turn slip lanes. With improvements to access at uncontrolled crossings, the number of visually impaired individuals who travel independently might increase.
Roundabouts are circular intersections with specific traffic control and design features. These features include yield control at entry, channelized approaches, and geometric approach curvature (deflection) to induce entering traffic to slow to the design speed of the circular pathway. The typical location for crosswalks at roundabouts is illustrated in figure 1. Note that one end of each crosswalk terminates on a splitter island (a median) so that the pedestrian crosses only one direction of traffic at a time. Because the crosswalks are set back from the intersection, conflicts with turning vehicles are minimized. Also, because all traffic is forced to travel on a curved path, vehicle speed is minimized, typically to between 32 kilometers per hour (km/h) (20 miles per hour (mi/h)) and 44 km/h (30 mi/h). Additional information on the design of roundabouts may be found in the Federal Highway Administration (FHWA) publication Roundabouts: An Informational Guide. (6)
Roundabouts are appearing with increasing frequency in the United States, primarily because of their benefits to traffic operations and safety. Roundabouts usually result in less traffic delay than a signalized intersection, and they may have operational benefits as alternatives to intersections that are not signalized. (6) Numerous studies have shown that roundabouts have fewer injury-related crashes when compared to conventional intersections, and most studies also show that the overall crash rate is reduced when conventional intersections are replaced with roundabouts.(6,7)s Furthermore, pedestrian death and serious injury rates are lower at roundabouts than at conventional signalized and nonsignalized intersections. (6)
Motorists' Yielding Behavior
It has been widely observed that motorists frequently fail to yield to pedestrians in crosswalks, even though most motorists are aware of this legal requirement.(8,9) For instance, at five uncontrolled crossings in Madison, WI, the percentage of vehicles yielding to pedestrians who were starting to cross ranged from 0 to 10.6 percent. (10) Huang et al. reported greater variation in the percentage of pedestrians to whom approaching motorists yielded. (11) Their results, from 11 different uncontrolled crossings in 4 States, ranged from 0 to 87 percent, with a mean of 50 percent. Huang and Cynecki also reported on the percentage of pedestrians yielded to by approaching motorists. (12) Their study included data from eight uncontrolled crossings in seven States. Yield rates ranged from 0 to 58 percent and averaged 19 percent.
Street Crossing by Visually Impaired Pedestrians
The U.S. Access Board Web site provides a detailed description of the process a visually impaired person would typically engage in to prepare for and make a street crossing. (1) Although this process has many implications for roundabout accessibility, the present study addresses only one of these implications: the need to provide detectible gaps in traffic. A gap is any pause in traffic flow during which a pedestrian can cross a roadway without encountering vehicle conflicts. At signalized crossings, visually impaired pedestrians rely on the signals to create gaps, and may use surges in traffic flow to detect the signal phase. At uncontrolled crossings, such as at roundabouts, the visually impaired cannot predict gaps, and they must rely on other environmental cues to detect gaps. For a visually impaired pedestrian to make a safe crossing at a roundabout, a gap in traffic must occur; the pedestrian must be able to recognize when gaps are present; and the pedestrian must be able to recognize when gaps are not present.
Gaps may occur in two ways:
For gaps to be usable by pedestrians, the gap must be of sufficient length, and the pedestrian must be able to detect them. Sighted pedestrians detect gaps by looking in the direction of traffic. Visually impaired pedestrians detect gaps by listening. The visual and auditory information of approaching traffic can be blocked for either sighted or visually impaired pedestrians by obstructions such as landscaping, hills or bends in the road, and buildings. Although in certain circumstances the approaching traffic can be heard before it is seen, more often the sighted pedestrian can detect the approaching traffic long before the traffic can be heard, even under ideal listening conditions, because of features that affect the acoustics such as road surfaces and reflecting surfaces. Although visual information for sighted pedestrians can sometimes be obscured by fog or glare, the sounds upon which visually impaired pedestrians rely are much more easily masked by other sounds. Masking sounds often include the noise of receding vehicles or traffic at nearby intersections, but they can also come from many other environmental sources. (13, 14, 15)
Thus, there are many situations in which sighted pedestrians can easily determine gaps, but visually impaired pedestrians are unable to reliably identify the same gaps. Whereas sighted pedestrians may be able to see gaps behind approaching vehicles, the visually impaired cannot hear such gaps. Furthermore, because the sound of receding vehicles masks gaps, the visually impaired may not recognize gaps for several seconds after they begin.(3) At a double-lane roundabout with 3.7-meter (m) (12-foot (ft)) lanes, a sighted person who walks at 1.2 meters per second (m/s) (4 feet per second (ft/s)) might accept a gap of little more than 6 s; however, the gap required by a visually impaired pedestrian must be larger. Vehicles passing the crosswalk create noise that can mask the silence of gaps behind them, and Guth et al. observed that it frequently takes 3 s or more before the noise of vehicles that have passed is diminished enough to allow detection of a following gap.(3) Thus, a visually impaired person may require a gap of 9 s or more. Furthermore, although the splitter island may represent a protected location for a sighted person to cross by looking in one direction at a time, the sound of traffic approaching and receding on the other side of the island is not masked by the island. Thus, the visually impaired pedestrian may require a 9 s gap in both directions before beginning to cross. After reaching the splitter island, visually impaired pedestrians may again require 9 s gaps in both directions because the sounds of traffic behind them will mask gaps in the entrance lanes.
Given that simultaneous gaps of 9 s or more at both entry and exit lanes may be required for detection of naturally occurring gaps in moving traffic, and that gaps such as these may be rare during peak traffic at double-lane roundabouts, it may be necessary to stop traffic to provide access to visually impaired pedestrians. The U.S. Access Board recognized this when it drafted a proposal to signalize roundabouts. (4) Signals might not be necessary if drivers would stop for pedestrians with long white canes or guide dogs, and if those pedestrians could detect the resulting gaps in stopped traffic. The purpose of the present study was to examine treatments that might make this latter approach feasible.
The goals of the present studies were to: (1) evaluate a method of aurally indicating when vehicles have yielded in both lanes of a double-lane roundabout exit (2) evaluate a method for increasing motorists' yielding to visually impaired pedestrians, and (3) quantify motorists'-yielding behavior at double-lane roundabouts.
Study 1 addressed only the first of the three goals: to evaluate a method of aurally indicating when vehicles have yielded in both lanes of a double-lane roundabout. Not all visually impaired pedestrians will cross in front of idling vehicles; without the benefit of eye contact with the driver, it is difficult to determine a driver's intent or the reason that the driver stopped in the first place. To cross a single lane, the visually impaired pedestrian listens for the idling engines of stopped vehicles. Some visually impaired pedestrians will cross in front of stopped vehicles. Crossing two lanes is more problematic than crossing one lane, because the pedestrian will normally want to determine that drivers have stopped in both lanes. However, the sound of one idling vehicle may mask the sound of another approaching vehicle, and the driver of that vehicle may not be looking for (or be able to see) a pedestrian stepping out from in front of a previously stopped vehicle. Unfortunately, as the results of this study show, the sound of one idling vehicle may also mask the sound of a second idling vehicle, especially when the first vehicle is in the lane closest to the pedestrian.
In the treatment condition of this study, strips were placed on the pavement to generate a sound that could be heard over the sound of idling vehicles and other traffic. The strips were needed to: (1) alert the pedestrian when vehicles approached in each of two lanes; (2) indicate when a vehicle passed the crosswalk (i.e., departed); and, most important, (3) indicate when an approaching vehicle did not pass and, by implication, was stopped. To ensure that all of the conditions of interest could be tested effectively, this study was conducted on a closed course where drivers yielded, failed to yield, and departed according to a script.
Study 2 addressed all three goals (listed above). The study was conducted at a double-lane roundabout in Maryland. As with study 1, the effectiveness of sound-generating strips was evaluated. In addition, two traffic control signs were deployed to evaluate their effect on motorists' yielding behavior.
One of these signs was placed at the upstream edge (from a vehicle perspective) of the crosswalk on the demarcation between the travel lanes. This sign, designated R1-6 in the Manual on Uniform Traffic Control Devices (MUTCD), was 90 centimeters (cm) (36 inches) high by 30.5 cm (12 inches) wide, with the sign bottom mounted 61 cm (23 inches) above the pavement on a breakaway post.(16) The sign, depicted in figure 2, shows a red triangular yield icon above the word "TO," which is above a pedestrian icon, all on a white (i.e., regulatory) background. The white background is surrounded by yellow with the words "STATE LAW" above and the words "WITHIN CROSSWALK" below. Combining the words and icons in English, the sign reads, "State law, yield to pedestrian within crosswalk."
The second sign was placed below an existing crosswalk warning sign located to the right of the crosswalk. This white sign, designated R1-5 in the MUTCD (depicted in figure 3), had an arrow that pointed, roughly, to the upstream crosswalk demarcation line, a yield symbol, the words "HERE" and "TO," and a pedestrian icon. By combining the words and icons, the sign reads, "Yield here to pedestrian." The sign was 61 cm (24 inches) square. The purpose of adding this sign was to encourage drivers to stop at the crosswalk rather than at a distance away from the crosswalk, which might prevent the pedestrian from hearing idling engines.