There are preliminary results of a test implementation of the Google Green Light project, which is based on machine learning and artificial intelligence and with which the Internet company models traffic patterns and suggests optimized traffic light switching. Thanks to the technology, traffic on some of Seattle’s busiest streets is running a little more smoothly, reports Scientific American. A spokeswoman for the transportation department in the US West Coast city quoted Science magazine about positive experiences. Green Light made “concrete, actionable recommendations”, drew attention to bottlenecks in the transport system and confirmed known bottlenecks.
Advertisement
Google launches its Green Light pilot program in Seattle and a dozen other cities in late 2023. In addition to some notoriously congested metropolises like Rio de Janeiro in Brazil and Calcutta in India, this also includes Hamburg. During these trial runs, local traffic engineers use the system’s suggestions to adjust traffic light switching. With this initiative, the company aims to reduce waiting times at traffic lights, improve traffic flow on busy main roads and intersections, and ultimately reduce greenhouse gas emissions. According to Google, early figures show the potential for up to 30 percent fewer stops at traffic lights and 10 percent fewer CO2 emissions there.
the origin of green light There is an AI-based model of each intersection. This includes their structure, traffic patterns such as driving and stopping periods, the traffic light scheme and the way traffic and signaling systems interact. Based on this pattern, the system develops recommendations that can be transmitted to city planners, for example, via a special interface. A big advantage of the project: it does not require expensive, permanently installed sensors and does not need to constantly check on site. Instead, it compiles existing traffic data from Google Maps collected from vehicles and smartphone users, ultimately acting as “mobile sensors”.
Greenlighting proposals sometimes has no direct impact
Henry Liu, head of the Transportation Research Institute at the University of Michigan, has a more cautious view of the technology. According to him, Green Light in Birmingham was able to reduce the time spent at intersections by 20 to 30 percent. “It all depends on the basis of comparison,” Civil and environmental engineer explains to Scientific American. In Birmingham, for example, only traffic lights have fixed hours. These were based on the number of cars which had not been updated for a long time. The climate argument should also be viewed with caution: loud Official information Waiting cars and traffic jams account for only two percent of all traffic-related emissions in the United States. Anyone driving at a higher speed Consumes quite a lot of fuel Like a car stopped because of a red light.
Green Light does not yet take into account complex factors such as bus and bicycle lanes, trams and busy pedestrian crossings. Still, the tips are not always successful. In Seattle, planners withdrew a change to the traffic light switch because it ultimately had no benefit. In another pilot city, Manchester, traffic engineers repeatedly chose to deliberately ignore Google’s recommendations. According to them, traffic lights there are sometimes deliberately placed so that buses get priority or commuters in residential areas have to spend more time. The AI approach of reducing the number of stops at intersections is counterproductive here. In this country, for example, the AI traffic lights in Essenbach and Hamm, which work with the system from Unex Traffic, still cause considerable annoyance to drivers.
(Never)
