Skip to contentUnited States Department of Transportation - Federal Highway Administration FHWA Home
Research Home
Report
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

 

Alternate Text for HRT 05-080

Pedestrian Access to Roundabouts: Assessment of Motorists'
Yielding to Visually Impaired Pedestrians and Potential Treatments To Improve Access

Figure 1. Drawing. Illustration of single–lane roundabout with crosswalks.

This figure depicts a four–legged roundabout intersection of two two–lane roads, where the roads intersect at right angles to each other. Unlike a traditional intersection, all lanes flow into and out of a single–lane circular roadway—the roundabout—in which vehicles enter and travel counterclockwise until they reach the desired intersection leg. The circular roadway is labeled "circular roadway." Yield lines, shown as dashed lines, mark the borders between entry lanes and the circular roadway. On each leg a splitter island, in the form of a raised median, is shown between the entry lane and the exit lane. Marked crosswalks are shown on each leg. The crosswalks are about a car length away from the circular roadway. The crosswalks converge on the splitter islands.

Figure 2. Picture. MUTCD RI—6.

This figure shows the "yield to pedestrian within crosswalk" sign that was placed between traffic lanes in the field study. The sign is 36 inches tall by one foot wide. The outer border of the sign is yellow. At the top of the sign, black letters on a yellow background read "STATE LAW." At the bottom of the sign, in smaller black letters on the yellow background, are the words "WITHIN CROSSWALK." The central portion of the sign has a white background. At the top of the white area is a red yield symbol-a red, nearly equilateral, triangle with a point at the bottom and a flat side at the top. Within the red triangle is the word "YIELD" in red letters against a white background. Directly below the yield symbol is the word "TO" in black letters. Below this word, still in the white rectangle, is a pedestrian icon. Combining the words and icons, the sign reads, "State law, yield to pedestrian within crosswalk."

Figure 3. Picture. MUTCD RI—5.

symbol, the words "HERE" and "TO," and a pedestrian icon. Combining the words and icons, the sign reads, "Yield here to pedestrian."

Figure 4. Drawing. Layout of closed–course test facility.

This figure shows an engineering drawing of the closed test course. The course is rectangular, with curved corners that have large radii. The experimental crossing is on the left side of the course. The location of the platform where participants stood in the experimental condition is shown juxtaposed to the outside edge of the course. The platform where participants stood in the control condition is at the bottom edge of the course. Arrows indicate that the vehicles traveled in a counterclockwise direction around the course. The letters D, C, B, and A are arranged around the top of the course. These letters indicate where vehicles were staged between trials. Stations D and B are on the inside lane, and stations C and A are on the outside lane. The scale on the drawing shows that the approximate diameter of the course was 61 meters, or about 200 feet.

Figure 5. Form. Example of driver's script.

The trial is identified at the top of the form with the title Trial 16. It also states begin from station C, followed by yield immediately. The table on the form has four columns for vehicles 1, 2, 3, and 4 and three rows for laps 1, 2, and 3. For example, using the table, the driver of vehicle 3 knows to start at station C. The instructions are in bold and for lap 1 say right first; for lap 3 the instruction is station A. A key below the table tells the driver that right first means yield (stop for pedestrian). The other actions in the key that drivers may take include:

Go: drive past the pedestrian without stopping.
Go slow: drive slowly, less than 5 miles per hour, pas the pedestrian.
Yield: stop for the pedestrian.
Stop: Stop behind a car already stopped at the crosswalk.
Left first: Yield, stop for pedestrian.
Left second: Yield, stop for pedestrian.
Right first: Yield, stop for pedestrian.
Right second: Yield, stop for pedestrian.
Vehicle instructions not in bold indicate that it will not move during the trial.

Figure 6. Graph. Proportions of correct identifications of stopped vehicles shown by test condition, the lane in which the first vehicle to yield stopped, and the lane identified.

This graph shows the proportions of correct identifications of stopped vehicles by lane (far or near), which lane yielded first (far or near), and experimental treatment (treatment or control). From top to bottom and left to right, the graph shows the following:

In the treatment condition, when a vehicle was already stopped in the far lane, correct identification of vehicles stopping in the near lane was 98 percent. Also in the treatment condition, if the vehicle in the far lane was first to yield, it was correctly detected 81 percent of the time. In the treatment condition, when the vehicle in the near lane was first to yield, it was correctly detected 73 percent of the time. In the treatment condition, vehicles that yielded in the far lane when a vehicle was already stopped in the near lane were correctly detected only 54 percent of the time.

All the following apply to the control condition: Vehicles that stopped in the near lane when a vehicle was already stopped in the far lane were correctly detected 71 percent of the time. Vehicles that yielded first in the far lane were correctly detected 64 percent of the time. Vehicles that yielded first in the near lane were correctly detected 68 percent of the time. Vehicles that yielded in the far lane when another vehicle was stopped in the near lane were correctly detected only 31 percent of the time.

Figure 7. Photograph. Roundabout exit in baseline configuration. This is a photograph of the crosswalk used in study 2. The view is from the sidewalk, approaching the exit from the roundabout side. To the right in the picture is the sidewalk and wheelchair ramp to the crosswalk. The crosswalk is in the center of the picture. The pavement can be identified as asphalt. The width of the crosswalk is demarcated by two 0.3–meter concrete strips on either edge that are transverse to the roadway. Between the concrete strips is a brick sidewalk that is about the same width as a car that is shown driving through the crosswalk. The splitter island and far–side crosswalk are shown in the left center of the image. Houses and trees in the background suggest a suburban neighborhood location. In the center of the picture (on the far side of the street), the mast on which the cameras used in the study were mounted is visible. The mast is a white vertical pole just to the left of a pickup truck that can also be seen in the background.

Figure 8. Photograph. Sound–strip installation and mount for the street sign. This photograph of the crosswalk used in study 2 was taken from the sidewalk from a perspective approaching the roundabout. The pavement treatment strips can be seen, though not clearly. The street sign that was used in the study is not shown, but the mount to which that sign was attached can be seen at the edge of the crosswalk on the roundabout side. The mount is positioned where the white line that demarcates the two vehicle lanes intersects with the edge of the crosswalk. The central island of the roundabout can be seen in the background. It is covered with plants and grass and two small trees that have shed their leaves.

Figure 9. Graph. Distance from crosswalk of vehicles stopping in the near lane. This is a bar chart graph that shows the number of vehicles that stopped at various distances from the crosswalk. The data shown are:

1.5 meters, 24 vehicles
3 meters, 19 vehicles
4.5 meters 11 vehicles
6 meters, 41 vehicles
a dashed line at about 6.1 meters indicates the distance from the crosswalk for one of the sound strips.
7.5 meters, 26 vehicles
a dashed line at about 7.4 meters indicates the location of another of the sound strips
9 meters, 37 vehicles
10.5 meters, 28 vehicles
12 meters, 1 vehicle
13.5 meters, 4 vehicles
16.5 meters, 3 vehicles
All distances are from the outside edge of the crosswalk to the front bumper of the car. Distances are approximate and estimated from recorded video images.

Distances are approximate and estimated from recorded video images.

Figure 10. Photograph. Two vehicles that stopped for the pedestrians. This is a video capture that show the COMS, participant, and guide dog (not seen) as they are about to cross. Two vehicles have stopped for them, the only two vehicles in the picture. A light–duty truck is stopped in the near lane about 7.5 meters away from the crosswalk. The truck's bumper is almost at the same distance as the sound strip that is farthest from the crosswalk.

The car stopped in the far lane is a car length behind the truck, perhaps five car lengths from the crosswalk or about 16.5 meters.

Figure 11. Graph. Distance between the crosswalk and a vehicle stopped in the far lane is shown as a function of treatment condition.

This bar chart shows the number of vehicles that stopped various distances from the crosswalk in the far lane. Separate bars are shown for the control and treatment conditions. The frequencies are listed here in the following order: the midpoint distance (e.g., 1.5 meters includes all vehicles that stopped distances between 0.75 meters and 2.25 meters), the frequency for the control condition, and the frequency in the treatment condition. The mode for the control condition is 11 meters; the mode for the treatment condition is 17 meters.

The data points are:
1.5 meters, control 3, treatment 1
3 meters , control 2, treatment 3
4.5 meters, control 1, treatment 3
6 meters, control 8, treatment 0
7.5 meters, control 2, treatment 3
9 meters, control 5, treatment 1
10.5 meters, control 9, treatment 5
12 meters, control 1, treatment 3
13.5 meters, control 5, treatment 6
15 meters, control 1, treatment 4
16.5 meters, control 1, treatment 10
18 meters, control 0, treatment 0
19.5 meters, control 0, treatment 2

Figure 12. Photograph. Pedestrians crossing after a correct detection of both lanes blocked.

This video capture shows the participant and mobility specialist entering the crosswalk. A car that has stopped within a meter of the crosswalk blocks the right lane. Six vehicles are stopped behind that car, including a transit bus that is turning right into the circular roadway from the upstream entrance. A single vehicle that is stopped in the circular roadway about 12 meters from the crosswalk blocks the left lane.

Figure 13. Histogram. Frequencies of delay between the times both lanes were blocked by stopped vehicles and correct detection of that state by participants.

This is a frequency histogram of the amount of time in seconds that it took participants to correctly identify that both lanes were blocked by stopped vehicles. Below are two columns of numbers. The first column shows the number of seconds (from negative 2 to 9) in bins of 1 second. The second column shows the number of identifications that were recorded for each of the durations given in the first column. Thus, in the first row, negative 2 3 means that there were 3 trials in which participants correctly indicated that both lanes were blocked 2 seconds before both vehicles came to a complete stop.

Negative 2 seconds with a corresponding frequency of 3
Negative 1 seconds with a corresponding frequency of 0
0 seconds with a corresponding frequency of 4
1 second with a corresponding frequency of 1
2 seconds with a corresponding frequency of 1
3 seconds with a corresponding frequency of 5
4 seconds with a corresponding frequency of 4
5 seconds with a corresponding frequency of 1
6 seconds with a corresponding frequency of 1
7 seconds with a corresponding frequency of 0
8 seconds with a corresponding frequency of 1
9 seconds with a corresponding frequency of 2
10 seconds with a corresponding frequency of 0

Figure 14. Histogram. Amount of time drivers waited before moving on.

This figure shows a histogram for the amount of time drivers remained stopped for the pedestrians if the pedestrians did not begin to cross. The data for the histogram are given in 3 columns and 9 rows.

The first value in each row is the longest time vehicles in that bin waited. The second column is the number of vehicles in the control condition in that bin. The third column is the number of vehicles in that bin from the treatment condition. For example, the first row contains the numbers 3, 1 and 21. This indicates that for durations as long as 3 seconds there was one vehicle in the control condition and 21 vehicles in the treatment condition. In the second row, the bin represents the number of vehicles that waited more than 3 seconds and up to 6 seconds. There were 5 such vehicles in the control condition and 20 in the treatment condition. The results across the graph were:

Wait in seconds 3, control count 1, treatment count 21
Wait in seconds 6, control count 5, treatment count 20
Wait in seconds 9, control count 6, treatment count 6
Wait in seconds 12, control count 5, treatment count 6
Wait in seconds 15, control count 3, treatment count 2
Wait in seconds 18, control count 5, treatment count 0
Wait in seconds 21, control count 2, treatment count 1
Wait in seconds 24, control count 0, treatment count 0
Wait in seconds 27, control count 1, treatment count 0

 

FHWA-HRT-05-080

Table of Contents

ResearchFHWA
FHWA
United States Department of Transportation - Federal Highway Administration