Green lights all the time: how traffic light assistance systems will improve traffic flow

Traffic light assistance systems let vehicles drive through cities more efficiently, saving electricity and gas. To make it work, more cars need to be networked.

Every driver has probably imagined a life without traffic lights at some point. But what if traffic flow actually improved? If we cannot do without traffic lights, what about zipping through a series of phased green lights? Traffic light assistance systems should make this possible. In the United States, some production cars are already communicating with traffic lights.

One of the first manufacturers to provide vehicles capable of this is Audi, which connected its cars with the city of Las Vegas. At the start of the project, Head of Electronics Pre-Development at Audi Andreas Reich announced: “For the first time, our cars are exchanging data with traffic infrastructure in real time”. The city provides traffic light data, which is processed and then sent to the vehicle via a fast internet connection.

The cockpit then shows the driver if the vehicle will make it through the next light before the phase changes. If not, the so-called “time to green” countdown begins. Audi believes this discourages drivers from accelerating to make the light.

The traffic-light system as a “byproduct” of automated driving

With this new kind of “anticipatory” driving, fuel consumption can be reduced by up to 15 percent, says Michael Zweck, project manager for Audi Traffic Light Information. The service is still limited at the moment. It is only available in the newer A4 and Q7 models, and only in Las Vegas and Washington, D.C. Audi sees the lack of unified traffic technology, which exists in Europe for historical reasons, as an obstacle to a rapid introduction of the technology there.

Whether in the US, Germany or elsewhere: to make such service available nationwide, “car manufacturers, local authorities and traffic control centers would have to adopt uniform technologies”, says Michael Crusius, spokesman for technology at Audi. He believes the amount of work necessary (including traffic light conversions) is too great just for implementing a single assistance system. Yet he has no doubt that traffic-light assistance systems will be commonplace in the future. As soon as cars are fully automated, the networking of traffic lights will follow.

This is because cars will certainly have to process traffic-light data when they are driving on the road without any human intervention. “Traffic-light recognition is necessary for autonomous driving,” says Georg Walthart, spokesman for technology issues at Mercedes, which it is also researching traffic-light assistance systems.

Swarm intelligence of networked cars: Car-to-X is necessary

Some cars already indirectly warn each other of black ice, temporary speed limits, or a broken-down vehicle behind the next hill. Yet this communication only takes place between vehicles of the same brand. The manufacturers see these functions as the start of Car-to-X communication. This means the networking of vehicles through infrastructure, in this case through manufacturer cloud servers where the data is analyzed. Swarm intelligence resulting from the networking of vehicles – when data from many vehicles is compiled to form a big picture, such as the traffic situation – is considered to be a fundamental requirement for autonomous driving

The acquisition of the Nokia Here mapping service by Daimler, Audi, and BMW in 2015 – the purpose of which was to help create the high-precision maps necessary for automated driving – is not surprising. If accidents are to be avoided, cars must be able to rely on centimeter-precise information. The HERE platform is therefore accessible in all sectors and can be used by car manufacturers, tech companies, and service providers.

Traffic-light assistance enables phased green lights for networked cars

In addition to buildings, curbs, guardrails, and the slopes and courses of roads, details about the surrounding technical infrastructure – traffic lights, for example – are among the data incorporated into dynamic maps, which are updated constantly in real time.

Once traffic-light assistance systems are commonplace, phased green lights can be integrated into route guidance, for example. More predictable stops at traffic lights could enable electric cars to more effectively charge the battery from regenerative braking while rolling to a stop.

“In 20 seconds, the light will turn green, so I’ll drive 30 instead of 50 so I don’t have to stop.” Such scenarios not only mean increased efficiency, but also more comfort for the driver, says Walthart. “The situation where the driver behind you honks because you missed the right moment will be a thing of the past,” the Mercedes spokesman says.

His colleague at Audi predicts that red-light violations and rear-end collisions will also become more seldom. Optimal traffic safety will become possible only after fully automated driving establishes itself. This is because the robot car, unlike human drivers, will not take any risks.