Traffic jam no more?: How smart traffic lights fight congestion

As cities grow larger and larger, the need for sufficient transport keeps rising steadily. This causes severe problems of overload: When too many people use public transport, some simply won’t fit into the trains and trams available and those who do certainly won’t have a pleasant ride. Even if you go by car, the experience can become harrowing. In grid-based cities, congestion leads to grid lock, making it impossible to move at all. Stuck in traffic, the minutes go by, your nerves wear thin, and you idly sit there until the traffic gods allow you to move on. These are seconds that are wasted pointlessly every day. And this problem scales, so in the big picture, there are thousands of hours chucked to the curb every day while waiting for the light to go green again.

Smart traffic light systems are tackling this problem and aim to cut back traffic light waiting time. By utilising cameras and sensors that feed computer models, intelligent traffic management can modify the times a traffic light stays red or green to make traffic more efficient. If there is a sudden surge in vehicles passing an intersection from one direction, the system can adapt and set longer green intervals to allow more vehicles to come through.

Introducing smart traffic lights is a measure that can help to get a grip on several problems at once. Fewer hours spent waiting for the light to turn green means fewer litres of petrol burnt and fewer emissions emitted. It also means less time wasted that can be used for work or recreation instead.

This concept does not only work for individual junctions. It can be scaled, so the systems of all traffic lights are interconnected, resulting in the highest possible throughput. If the bulk of cars crosses a junction westward, the following lights in that direction can adapt and allow vehicles going into that direction to pass by green lights, thereby not requiring them to stop. This turns a city’s fragmented intersections, into a net of nodes that are constantly in contact with one another, ready to change their pattern when necessary. By applying game theoretical concepts and algorithms, the importance of every single node in the system is ranked and a sequence is internally calculated that brings the best possible outcome. If congestion under this system does occur (maybe because of an accident) the lights surrounding the site can factor in the obstruction in their intervals, allowing the congestion to dissolve.

A predecessor to the much more complex traffic management systems taking shape today, has been the installation of sensors or magnetic induction loops into the asphalt that can perceive vehicles at an intersection. If a vehicle triggers the system, it reacts by turning green sooner. It basically works like a pedestrian light on a busy street that only turns green when a button is pressed. Although this approach surely constituted an improvement, it only saved a few seconds per light phase at most. Because the system could not predict when traffic waves would come, it only saved time when vehicles had already come to a halt.

In order to test and improve smart traffic light systems before they are rolled out more broadly, projects in many cities around the globe have taken shape. One such system is SURTRAC from Pittsburgh, which is running since 2012 and has been continually expanded. As of 2016, SURTRAC has been deployed at 50 intersections along heavy-traffic roads in the eastern Pittsburgh area. The project developed at Carnegie Mellon University shows impressive results, cutting travel time by 25 percent and idle time by over 40 percent.

A predecessor to the much more complex traffic management systems taking shape today, has been the installation of sensors or magnetic induction loops into the asphalt that can perceive vehicles at an intersection. If a vehicle triggers the system, it reacts by turning green sooner. It basically works like a pedestrian light on a busy street that only turns green when a button is pressed. Although this approach surely constituted an improvement, it only saved a few seconds per light phase at most. Because the system could not predict when traffic waves would come, it only saved time when vehicles had already come to a halt.

In order to test and improve smart traffic light systems before they are rolled out more broadly, projects in many cities around the globe have taken shape. One such system is SURTRAC from Pittsburgh, which is running since 2012 and has been continually expanded. As of 2016, SURTRAC has been deployed at 50 intersections along heavy-traffic roads in the eastern Pittsburgh area. The project developed at Carnegie Mellon University shows impressive results, cutting travel time by 25 percent and idle time by over 40 percent.

Even further into the future, development of communication between vehicles and infrastructure could spark the replacement of traffic signs and lights with on-board signalling that shows motorists when and how to pass intersections individually. This would make it possible to adjust signals more quickly and efficiently, as safe corridors can be allocated to every single vehicle separately, not to a whole group of vehicles going a specific way. The logical conclusion for further improvement would be the implementation of such a system into driverless cars, creating a closed, adaptable and autonomous traffic system.

Meanwhile in the present, we are still stuck with some rather straightforward issues like people crossing red lights. In order to save a few seconds on a drive, some are prepared to accept potentially deadly risks. Even when discouraged by higher fines, offenders will take a chance when they feel a red light stays on too long pointlessly. But when light phases get shorter or simply more efficient, it gives people less reason to do so. With the help of smart traffic lights, drivers might be less inclined to cross a red light, when the alternative is waiting for only a moment’s time.

But what if lights intentionally delay drivers? To stop people from violating the speed limit, the town of Swindon in the UK has installed speed traps that can initiate the next traffic light of offenders to turn red. This way, the trespasser has to sit out a few seconds of punishment for going too fast. Hopes are, that this mechanism has potential speeders to make up their mind beforehand, taking it a little more slowly instead of waiting at the light.

Not all developments for smart traffic management are necessarily an improvement. Authorities in Augsburg, Germany and Sydney, Australia, have installed special traffic lights directly into the ground, attempting to warn people whose eyes are glued too much to their phones – even when they are crossing the street. If pedestrians now look at their phone while crossing, they immediately see red lights blinking without looking up. But what sounds like a good idea can easily backfire and reinforce such dangerous behaviours. When the notion that it isn’t necessary to look up while crossing the road is affirmed, it puts pedestrians in even more danger.

Equipping traffic lights with smart computer systems has not been universally greeted with welcome. For one thing, as many systems use CCTV to monitor traffic flows, the institutional misuse of such infrastructure for surveillance purposes might be anything but a paranoid fantasy. This, however, only applies to camera-based systems; Sensor-based ones safer, since they can only collect anonymous data of how many vehicles move where, but not who the people driving or owning the vehicles are.

Making infrastructure smarter can obviously be a great tool to make cities run smoothly, but it also exposes severe vulnerabilities. When every traffic light acts independently, a software problem can take out that one

traffic light. But when all lights in a city are connected to each other, a bug or a hacker attack can take out the whole system, with grave consequences. Therefore, cybersecurity needs to be a top priority in the age of smart cities. If smart traffic management systems are implemented to be relied upon by millions of people, resolute security measures must be taken to ensure that they stay intact, even when people try to sabotage them.

Smart traffic lights are definitely a great technological development and their result do not lie. Saving tons of time, money, nerves and petrol is certainly an appealing outlook. But all the enthusiasm encircling the subject could use a little moderation, as smart traffic lights are no panacea. Trials of different systems show considerably different outcomes. Some bring significant improvements in traffic flow, a few barely any.

Even the best approaches can only postpone traffic meltdown in ever growing cities. They can only increase efficiency of the space used, but if vehicles run out of road to squeeze onto, the most efficient traffic management system will not prevent traffic jams. The only hope to satisfy the needs of megacities lies in greatly expanding public transport. Smart traffic lights are a first step in the right direction, but they are only a short-term solution to a long-term problem.