Audible Beaconing to Help Pedestrians Who are Blind

Cross Streets

Billie Louise (Beezy) BentzenJanet M. Barlow

David GuthAlan C. Scott

TRANSED 2012

Accessible Design for the Blind, 17/9/12, Slide 2

This project was supported by Grant #5 R01 EY12894-07 from the National Eye Institute, National Institutes of Health. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Eye Institute.

Accessible Design for the Blind, 17/9/12, Slide 3

A Far-side Walk Signal Gives Good Results at a Simulated Crosswalk

Compared simultaneous signals from both ends of crosswalk, signals alternating from one and to the other, and far-side signal,

16 m long (4-lane) crosswalk simulated in a parking lotAccuracy of crossing significantly better with far-side signal

Accessible Design for the Blind, 17/9/12, Slide 4

Problems with Far-side Walk Signal Beacons

Too noisy to be tolerated in a neighborhoodThe far-side walk signal for one crosswalk might be mistaken for the signal for another crosswalkThe far-side signal wouldn’t work if people on both ends of a crosswalk pushed the button to cross during the same cycle

Accessible Design for the Blind, 17/9/12, Slide 5

A Prototype Developed to Avoid the Problems

Comes on only in response to a button-press of 1 sec or longer—seldom heardComes on only for the crosswalk where the button is pushedUnlikely to have 2 pedestrians holding the button down for the same crosswalk during the same cycle

Accessible Design for the Blind, 17/9/12, Slide 6

Components of APS with Far-side Beacon

Accessible Design for the Blind, 17/9/12, Slide 7

Users’ Experience1. Users hear locator tone from pushbutton2. Users push and hold button for at least 1 sec3. Users hear alignment tone--7 repetitions of

locator tone audible from far-side speaker4. Users wait for walk signal—quiet locator tone

continues5. Users hear walk signal from near-side APS only,

and begin crossing6. When walk signal ends, users hear far-side

beacon (loud locator tone) during pedestrian clearance interval, while crossing is completed

Tested in Charlotte, North Carolina16 participants made total of 48 crossings with and without far-side beaconingBeaconing resulted in improvement in crossing within the crosswalk

Accessible Design for the Blind, 17/9/12, Slide 9

More Intersection Testing—Alpharetta, GA, Towson, MD, Austin, TX

Compared crossings using far-side beacon APS, standard APS, and tactile guidestripFar-side beacon and tactile guidestrip resulted in good accuracyParticipants enthusiastic

about far-side beacon

Accessible Design for the Blind, 1/10/2010, Slide 10

But is Far-side Beaconing Safe?

What if the intersection is not very wide?What if there are reflective buildings on the corners?What if two people at the intersection request a far-side beacon during the same cycle?

Accessible Design for the Blind, 17/9/12, Slide 11

Trying to Cause Trouble in Portland, OR

Characteristics of the intersectionCrossings were 2-5 lanes wide Buildings on all corners3 crossings without good parallel traffic cues3 crossings not well aligned with direction of approach5 crossings

Accessible Design for the Blind, 17/9/12, Slide 12

43rdAve & Sandy Blvd

Hancock Street

Sandy Blvd

43 Ave

Accessible Design for the Blind, 17/9/12, Slide 13

4 Crosswalks Used for Experiment

Hancock Street

Sandy Blvd

43 Ave

Accessible Design for the Blind, 17/9/12, Slide 14

Method

22 participants who had little or no vision made 2 round-trip crossings of each experimental crossing Each crossing had one of 3 beaconing conditions

Standard APS—no beaconingAPS with far-side beaconingAPS with far-side beaconing, with experimenter calling walk signal for another crosswalk— “confusion trials”

Accessible Design for the Blind, 17/9/12, Slide 15

Results for Portland Experiment

Accessible Design for the Blind, 1/10/2010, Slide 16

Standard APS--participants tended to veer out of the crosswalk, and did not successfully correct their heading; some diagonal crossings

Accessible Design for the Blind, 17/9/12, Slide 17

APS with far-side beaconing—participants often observed to correct their heading;

no diagonal crossings

Accessible Design for the Blind, 17/9/12, Slide 18

APS with far-side beaconing and “confusion”

No confusion noted

Accessible Design for the Blind, 17/9/12, Slide 19

Conclusions

APS with far-side beaconingEnable blind pedestrians to maintaining or correct their heading so they usually cross within the crosswalk

Accessible Design for the Blind, 17/9/12, Slide 20

Conclusions

APS with far-side beaconingEnable blind pedestrians to maintaining or correct their heading so they usually cross within the crosswalkDo not result in confusion about which crosswalk has the walk signal even when another person actuates beaconing for another crosswalk

Accessible Design for the Blind, 17/9/12, Slide 21

Conclusions

APS with far-side beaconingEnable blind pedestrians to maintaining or correct their heading so they usually cross within the crosswalkDo not result in confusion about which crosswalk has the walk signal even when another person actuates beaconing for another crosswalkCan safely be used at medium-sized intersections even where there are reflective buildings on the corners

Accessible Design for the Blind, 17/9/12, Slide 22

Recommendations

APS with far-side beaconing may be used to provide wayfinding information

At crossings where there is insufficient or inconsistent directional information from traffic, or very wide crossingsAt intersections where there are reflective buildings near the corners

Accessible Design for the Blind, 17/9/12, Slide 23

Cautions

APS with far-side beaconing Should be actuated by an extended button-press

Accessible Design for the Blind, 17/9/12, Slide 24

Cautions

APS with far-side beaconing Should be actuated by an extended button-pressDo not need to be programmed for all crossings at an intersection; e.g. short crossings or crossings with reliable parallel traffic

Accessible Design for the Blind, 17/9/12, Slide 25

Cautions

APS with far-side beaconing Should be actuated by an extended button-pressDo not need to be programmed for all crossings at an intersection; e.g. short crossings or crossings with reliable parallel trafficShould have especially careful attention to sound-level adjustment—no louder than necessary

Accessible Design for the Blind, 17/9/12, Slide 26

For additional information:

Contact Beezy Bentzen

bbentzen@accessforblind.org