Maryland State Highway Administration
APSs and Pilot Study
November 1999 Newsletter
Metropolitan Washington Orientation and Mobility Association (WOMA)
At our last meeting, we met with Tom Hicks, Director of the Office of Traffic and Safety for Maryland's State Highway Administration (SHA) to find out about their pilot study for Accessible Pedestrian Signals (APSs). We were pleased to see at the meeting Maryland SHA staff Linda Singer and Michael Niederhauser, as well as Bruce Mangum from the Traffic Management Center of Montgomery County's Department of Transportation.
There were four WOMA members present. Although we had invited the presidents of the Maryland chapters of the National Federation of the Blind (NFB) and the American Council of the Blind (ACB), neither were able to attend. However three members of the NFB came: Curtis Chong, coordinator of the NFB Technology Center in Baltimore; Tom Bickford, author of Care and Feeding of the Long White Cane, an NFB publication about traveling independently, and Jennifer Tyndaal, who had requested an APS at her neighborhood intersection (see our last newsletter). Also present were Margaret Buzzard, who had also requested an APS at an intersection near her, and Lois Thibault from the U.S. Access Board.
Mr. Hicks explained that he has been interested in the subject of APSs for about 8 years, when they installed a trial APS in Fredericksburg that sounded like a truck backing up. However he observed no consumer interest in APS until about 5 years ago. There are presently no standards for APSs, and he speculated that if there were indeed a need for APSs, there would have been standards by now.
He explained that engineers must follow the Manual on Uniform Traffic Control Devices (MUTCD), which is now being revised. Standards and guidelines for APSs are in the revised draft of the manual, which will go out for comment and voting before the final draft is ready next year.
Mr. Hicks also said that although the Maryland SHA is completing the APS pilot study, they will wait for national standards to be established before installing any more APSs. He said that it is important that there be uniformity and consistency in APSs before they are installed, just as there is uniformity in the traffic signals among the states (green means "go," and red means "stop").
Consistency would indeed be very helpful. For example, there is presently inconsistency in the pedestrian signals for sighted pedestrians, and it is confusing. In Maryland, it's hard to know which street the pedestrian button is for unless a sign is installed (which is often not the case) because sometimes it faces the street it is intended for crossing, and sometimes it is facing the other street. Sometimes the signal shows a hand and a man walking, sometimes it shows the words "stop" and "walk." Throughout the U.S., some places use LED displays, and experimental pedestrian signals show graphics of eyes looking toward the cars to remind pedestrians to watch for turning cars. Most jurisdictions have solid white walk signs, but Washington DC has a flashing white walk signal whenever the pedestrian needs to watch for turning traffic, and a solid white walk signal only when no traffic is allowed to turn in front of the pedestrian.
As Mr. Hicks points out, there should be consistency for pedestrian signals. As a sighted person, I am often confused as to which street the pedestrian button is for, and was confused when I first saw the flashing walk signs in DC. It could be confusing for DC residents traveling outside their city if they don't realize that outside of DC, solid pedestrian signals do not mean that no cars can turn in front of them.
So consistency among the states and municipalities would definitely be an improvement. However consistency is not required for the pedestrian buttons and walk signals for sighted pedestrians, so lack of consistency of APSs should not be a reason to withhold them from blind people.
Nevertheless, the AER Division Nine Environmental Access Committee shares Mr. Hick's concern about consistency. They consider it critically important that the APS information be unambiguous, and certain inconsistencies would make it ambiguous. For example the locator tone for the pedestrian button at one corner or in one city should not be confused with the sound of the walk signal at another. We applaud Maryland SHA and encourage it and others to work with neighboring and national jurisdictions to be sure there are no such potentially dangerous inconsistencies in APSs.
Mr. Hicks explained the pilot study briefly at the WOMA meeting, and also sent a letter to WOMA answering our questions more specifically. The information presented here is from both the meeting and his letter.
Maryland's SHA's APS pilot study originally involved three intersections along York Road in Baltimore. These were requested by three blind residents, all of whom were members of the ACB of Maryland. After the APSs were installed, the three people reported that the APSs gave information that may not be readily available regarding the appropriate time to cross; the proper crosswalk to use; and directional guidance to the upcurb. Some of their comments reflected a lack of understanding of how to use the signals. For example one of them noted that crossings could not be completed during the audible phase of the signal, apparently not realizing that the audible phase indicates when they can start walking, not how long they have to complete the crossing.
According to the letter, the pilot study was then expanded to seek input from the NFB, and later was expanded further to install APSs at intersections in Towson.
These later intersections were chosen because they offered alternate routes for getting around the Towson roundabout. However SHA did not feel (nor did consumer consensus indicate) that blind people needed APSs at those intersections to cross safely. This may explain why the responses and feedback that were obtained from the second stage of the pilot study did not specifically address whether the APSs were effective for resolving difficulties for crossing intersections ‑‑ no specific difficulties were determined to exist at the intersections where the APSs were installed, so it is not likely that any improvement would be noted.
Instead, the input seemed to be a survey of opinions about APSs in general. Input included statements that blind people do not need APSs to travel independently, concerns about the effect that APSs might have on the public's perception of blind people, and the possibility that blind people would feel unsafe crossing streets that don't have APSs or, conversely, that APSs would give blind people a false sense of security. This false sense of security apparently is the same false sense of security that sighted pedestrians experience with traffic signals. For example, one respondent to the pilot study asked what would happen if someone ran a red light where blind people relied on the pedestrian signal, and the same question could be asked about sighted people who rely on pedestrian signals.
The only comments which addressed specifics about APS and their use were two comments which pointed out that APSs might interfere with using traffic sounds. These comments were that the noise of the "chirping bird" covers up the traffic movement, and that APSs can make it difficult for blind people to listen to where the traffic is moving to determine when it is safe to cross.
This concern about APSs making it difficult to hear traffic was also raised by O&M specialists in a survey done last spring by the Environmental Access Committee of AER's Division Nine (see Division Nine's Summer 1999 newsletter for survey results). When Mr. Hicks finished his discussion, Dona Sauerburger, COMS presented information which included the Division Nine survey results, and showed a videotape of two intersections where blind people had difficulty or were unable to cross without an APS. She also demonstrated three APSs ‑‑ the Polara, the Prisma, and the Panich.
All three APSs addressed the concern of masking traffic sounds by being responsive to ambient sound. That is, when traffic is noisy, the APS sound is louder, and when it's quiet, the APS volume goes down.
In addition to responsiveness to ambient sound, all three APSs featured tactile arrows to indicate which street the button and signal were for, a locator tone which enabled the blind pedestrian to localize the button, and tactile as well as auditory indications of when the WALK signal is on. The Prisma also had a tactile graphic that illustrated geometric features of the street, such as the width, median strip, etc. The Polara verbally gave the name of the street being crossed if the button was held down for four seconds, and offered raised‑line and braille labels to name the street.
Dona explained that her experience matches that of Mr. Hicks, who had not heard any concerns from consumers until 5 years ago. Dona reported that until 5 or 10 years ago, her clients were able to cross almost all signalized intersections safely using traditional strategies. The reason that the need for APSs has increased since that time is because of more complex intersections (including actuation), quieter cars, wider streets, and more arterial streets which must be crossed to get to or from neighborhood entrances which provide little or no parallel traffic surges to indicate when the signal changes.
During the meeting, Bruce Mangum distributed a list of APSs which have been installed at intersections that are under the jurisdiction of Montgomery County. His statistics bear out what Mr. Hicks and Dona said: by 1995, there were only two APSs installed (1988 and 1991), but in the last 4 years, 5 more APSs were requested and installed. All but one of these were along arterials intersecting with quiet side streets, where there is insufficient parallel traffic surges from the side street to know when the walk signal is on. This is the same situation at the intersections where APSs were requested from Maryland's SHA by the two people in Dona's videotape and by Jennifer Tyndaal.
Strategies to Cross Streets at Traffic Signals Without APSs ‑‑ Is it Safe to Rely on Timing?
For years, WOMA and individual O&M specialists have struggled unsuccessfully to come up with alternative strategies to enable people to cross safely without APSs at modern intersections where traditional strategies aren't sufficient. We asked the NFB of Maryland, which opposes wholesale use of APSs, to help us figure out how blind people can cross certain intersections without them. We would also welcome ideas from the traffic engineers for how blind people can recognize when their pedestrian signal is on at difficult intersections where they refuse to install APSs.
In response to our request to the NFB, Tom Bickford graciously took the time to observe and assess a typically difficult intersection along an arterial, and proposed a strategy to cross without using an APS.
The intersection which he assessed was the one for which Jennifer Tyndaal had requested an APS. It is on Riggs Road, which is 7 lanes wide, and the problem was insufficient traffic from the side street. Mr. Bickford's solution was a variation of a strategy that has been tried occasionally by O&M specialists. This strategy is to time the traffic signal to predict when it's going to change.
In general, the strategy of timing the signal works like this. First, you observe that your walk signal or green light comes on at a certain time after another event, for example 17 seconds after the left‑turning traffic on the parallel street starts, or 25 seconds after the perpendicular traffic starts. Then, when the observed event occurs, you start counting seconds until you assume your signal starts, and you begin crossing.
Being able to use this strategy requires that:
1) the timing of the cycle is reliable and predictable;
2) you can consistently and accurately measure time (the longer the time being measured, the more difficult this is); and
3) you have the cognitive ability to understand the procedure and remember the required timing accurately each time you use it, regardless of how long ago you used it last.
The last two requirements can be met by many independent travelers who are visually impaired, although this strategy leaves out visually impaired people who have problems with memory or understanding complex procedures, or who cannot accurately and consistently measure the timing.
Nevertheless, let's look at the first criterion, to see if this might be a viable strategy for those who are capable of using it. That criterion is that the timing of the cycle must be reliable and predictable in order for timing strategies to be reliable.
This means that the cycle must either be fixed‑time or, if it's actuated, all possible factors that could affect the timing must be considered. Most city intersections are fixed‑time, but in the suburbs it is very risky to assume that a signal is fixed‑time, even if observations indicate that it is. For example if you assess an actuated intersection at 5:00 PM you might incorrectly conclude that it is fixed‑time because most actuated intersections appear to be fixed‑time during rush hour, and many intersections are fixed‑time during the day and actuated in the evening.
Even if a traffic engineer confirms that the signal is fixed‑time 24 hours a day, engineers can change signals from fixed‑time to actuation in a twinkling at any time, or they may leave it fixed‑time but change the timing of the different phases. When they make these changes, the engineers do not send out public notices, nor contact all blind travelers whose safety depends on this knowledge and who may assume that the signal is still fixed‑time or that the time of each phase is the same as it was.
Thus it is not advisable to have our safety depend on a timing system that relies on intersections being fixed‑time with the time of the phases never changing. Can we develop timing strategies at actuated intersections?
The situation for which Mr. Bickford was asked to find a solution was an actuated intersection, and he came up with a variation on the strategy of timing the signal. On the face of it, he seems to have taken into consideration all the factors that could affect the timing. He observed that the street being crossed, Riggs Road, had a minimum time for its green light, so that when the pedestrian button is pushed to cross Riggs Road, its light does not change until Riggs Road has its minimum time. This minimum time is 45 seconds. That is, if the pedestrian button is pushed less than 45 seconds after the light for Riggs Road is red, the light will not change again until Riggs Road has had at least 45 seconds of green.
Mr. Bickford observed that if a minute passes without the light for Riggs Road being red, the light for Riggs Road will turn yellow as soon as the pedestrian button is pushed, and exactly 10 seconds later, the walk signal to cross Riggs Road begins. It did this consistently the entire time he was there.
Using his strategy, then, the pedestrian would wait until Riggs Road had the green light for at least a minute, then push the pedestrian button and count ten seconds while preparing to cross. At the end of ten seconds, the pedestrian simply assumes that the walk signal is on, and starts to cross.
Before we can teach our students to trust timing strategies such as this one, we must assure that every possible factor is taken into consideration, and that there is no possibility that something could go wrong without the pedestrian realizing it. Unfortunately, this strategy does not consider every factor, and things could very possibly go wrong.
For example, Mr. Bickford's strategy requires that the pedestrians realize whenever the light for Riggs changes to red, because they must wait at least a minute after the light is red before they push the pedestrian button. But the signal can change without the blind pedestrian realizing it, as explained here.
After Riggs Road has had its minimum of 45 seconds of green light, whenever a car passes over the sensors on the side street, the light for Riggs Road turns to yellow and red. This happens even when the car is not coming from the side street, but rather is turning left from Riggs Road and cuts the corner so closely that it rides over the sensors. This happened several times while I was there with Jennifer. Rather frequently, then, even when there is no car waiting on the side street, the light for Riggs Road turns red.
This would be no problem if the traffic on Riggs stopped so the blind pedestrian would realize the light changed. She'd then know that she has to wait another minute before pressing the pedestrian button. However, if there is a gap in traffic on Riggs Road during the red light, no cars will stop, and the blind pedestrian won't realize the light for Riggs had changed to red and back to green again.
This could happen very easily. The light for Riggs is red for only ten seconds (unless there is more than one car waiting at the side street). So if it turns red during a gap in traffic on Riggs Road that is 10 seconds or longer, the blind person will never know the light had changed.
Because of traffic signals within a half mile in each direction, there are many gaps in traffic on Riggs Road that are much longer than 10 seconds, even during busy times of the day. During one 15‑minute observation of Riggs traffic on a weekday afternoon, there were 26 gaps ‑‑ almost two every minute. The gaps each lasted from 5 to 22 seconds, with the average gap being 13 seconds. Seventeen of the gaps were 10 seconds or longer, which is more than one such gap every minute.
If a car from Riggs Road drove over the sensors at the beginning of one of these gaps (which is very likely if the driver was waiting for a gap to turn left into the side street), the light would turn red for Riggs Road without the pedestrian even knowing it because there would be no cars on Riggs to stop. If the pedestrian pushed the button within a minute of that car driving over the sensors, the walk signal would NOT come on in ten seconds, it would come after Riggs Road had 45 seconds of green.
However because the pedestrian would be under faulty assumptions based on unpredictable features, she might start to walk across a seven‑lane street during a red light.
[2006 NOTE: Mr. Bickford wrote an article about his views regarding this meeting and the use of timing to cross at signals in the February 2000 issue of the NFB’s Braille Monitor]
This is just one scenario in which assumptions about the signal could be erroneous. There are a number of ways in which actuated signals become dysfunctional and therefore unpredictable. For example whenever the actuation wires in the street are cut during road repair or construction, the signal reverts back to fixed‑time until it is repaired again, which often takes months. During this time, Mr. Bickford's 10‑second strategy would not work because the green signal would not come on 10 seconds after the button was pushed, it would come on at regular intervals regardless of the button. The pedestrian relying on this strategy would have to be very attentive, intelligent and knowledgeable enough about traffic signals and actuation to understand all the possible ramifications and realize when the strategy will not be effective.
I think very few of us understand the workings of the signals enough to predict everything that could happen. I have studied, observed, and dealt with actuation for at least 5 years, and attended sessions with traffic engineers to try to understand how the system works, including the WOMA session with engineers several years ago. Yet every time I think I finally understand actuation and how it works, I am dismayed to find an exception to the rule, a situation which I had not predicted.
For example, at our WOMA meeting at Maryland's SHA last month, in response to a question from one of the blind travelers, the engineers explained that if you push a pedestrian button when the light to cross your street is green, the walk signal will not come on right away, it will come on during the next cycle. This is what I had been teaching my clients for years, until two years ago when I encountered an exception to the rule.
What I discovered is that this rule is not true if you are crossing a secondary street at a fully actuated signal (that is, where there are walk signals and pedestrian buttons for both streets). Sometimes in that situation, if you push the pedestrian button during a green light, the walk signal does come on immediately, not during the next cycle. When they were asked about it, the engineers at first said this wasn't the case, then reconsidered and realized that it was true.
What other exceptions to the rules exist which we haven't observed and which engineers may forget to tell us? How can we teach our students to use timing to predict when their walk signal will begin when the system is so complex, and there are unknown exceptions to the rules? And do our students need to be very intelligent and have degrees in traffic engineering in order to understand and remember all the rules and exceptions and predictions in order to guess what signal information is being provided to sighted drivers and pedestrians? There is too much possibility for error, either because of misunderstanding or not applying the rules correctly, or because of exceptions to the rules that we were not aware of.
We are very grateful to the NFB of Maryland, and in particular to Mr. Tom Bickford for trying to help us figure out how to teach people to cross safely without APSs. We also appreciate the expertise and consideration of the engineers from Maryland's State Highway Administration and Montgomery County. We are all working on the same goals ‑‑ safe, dignified, independent travel for people who are visually impaired.
However, in spite of all our combined efforts, resources, and expertise, we have not yet figured out how blind people can reliably get signal information in some situations without an APS.
Back to Home page