Don’t Push the Red Button: A Case Against Manual Pedestrian Detection in Urban Areas

摘要

Traffic lights in urban transport systems are often designed to take into account the situation on the road, i.e. detect vehicles and optimize road traffic flow. A wide variety of methods is employed to detect vehicles - from cameras through induction loops to laser radars. However, pedestrians are usually left with the rather antiquated method of pressing a button.Such an approach can generate a number of problems, especially if the light settings are based on the premise that manual and automatic activation of lights are equally effective, i.e. that every pedestrian will by default press the button at the moment s/he approaches the crossing. Such an assumption (one of several made when installing pedestrian buttons) is false due to a number of factors. These include the need for the pedestrian to: identify that the lights need to be activated (which is the case at only selected crossings), find a push button, walk up to it and activate it. Another assumption is that the pedestrian will wait until the light changes to cross the road, regardless of the situation on the intersection. While this has been shown to be a false premise in the past, it is particularly unlikely to hold true when the pedestrian sees that parallel traffic streams have a green light while s/he has a red one (due to not pressing a button or pressing it too late). Additional problems may stem from inadequate accessibility of the buttons, the insufficient number or noticeability thereof, or their being situated off the default (shortest) path.While pedestrian buttons are often treated as a default solution, manual detection is in fact a way of reducing the attractiveness and effectiveness of walking as a mode of transport. This may not be obvious, but it is enough to imagine a driver forced to go through the process delineated above (having to identify whether the lights will not change by themselves, if so - looking around for a way of activating them, moving to a given point where the lights can be activated, waiting for them to change, then going back to the track which was leading from the source of his/her journey to the destination) - at every intersection - to see that they are in fact a problem.The paper presents a discussion of approaches to the activation of pedestrian lights based on the situation in Poland, with particular focus on Warsaw, where field research was conducted. The problems outlined above are quantified and possible solutions thereto are suggested.