Good Thinking

AI traffic lights could reduce waiting times at intersections

AI traffic lights could reduce...
Workers install cameras and radar sensors at the intersection which was used as a model for the study
Workers install cameras and radar sensors at the intersection which was used as a model for the study
View 1 Image
Workers install cameras and radar sensors at the intersection which was used as a model for the study
1/1
Workers install cameras and radar sensors at the intersection which was used as a model for the study

Ordinarily, traffic lights change color in response to predetermined rules and induction loop sensors embedded in the road. According to new research, however, making those lights artificially intelligent may keep traffic flowing faster and smoother.

The study is being conducted as part of Germany's KI4LSA project – it's funded by the German Federal Ministry of Transport and Digital Infrastructure, and includes several partner organizations.

One of those partners, the Fraunhofer Institute for Optronics, System Technologies and Image Exploitation, recently installed high-resolution cameras and radar sensors at a busy traffic-light-controlled intersection in the city of Lemgo. Over time, this setup recorded the number of vehicles waiting for the lights to change, the amount of time that each vehicle had to wait, and the average speed at which the vehicles went through the intersection.

That data was then used to train a machine-learning-based computer algorithm. It virtually experimented with different light-changing patterns – which would continuously adapt to real-time traffic conditions – to see which ones would work best at minimizing wait times.

According to the computer simulations, the best AI-based patterns could improve traffic flow by 10 to 15 percent. That said, the researchers admit that the simulations may not correspond completely to real-life conditions. With that in mind, the algorithm will be used to run the traffic lights in the actual intersection over the next few months, and may be adapted based on the team's findings.

In the related KI4PED project, Fraunhofer and other partners are additionally looking at reducing wait times and optimizing crossing times for pedestrians making their way across intersections. Among other things, LiDAR sensors will be used to assess the walking speed of individual pedestrians, ensuring that they have sufficient time to safely cross the street.

Source: Fraunhofer

10 comments
10 comments
guzmanchinky
It just makes so much sense for the computer to monitor the situation with a camera instead of just a few in ground sensors.
Brian Beban
About time. Dumb lights have you standing half your commute times based on timing delays and activation sensors which aren't flexible. My car has more Ai than any lights telling me when to stop, steer, alert me to spacing, etc. I think a modern traffic system which can sense empty approaches could save millions in lost time and frustration sitting at dumb lights when there is no traffic.
christopher
The tech to fix traffic lights has been around for 30+ years, and over that time I've seen it deployed exactly zero times.
Nobody seems to care about the time and pollution of stupid traffic lights, least of all anyone who installs them.
I suspect the problem relates to "safety" as well as apathy though - compliance for traffic control systems must be a major expense and pain...

However - invent a camera which can automatically fine people for driving through red lights at deserted intersections - and YAY - *those* get installed EVERYWHERE...
Ben Grillet
Yeah, or, rip out ALL the stupid, inefficient, expensive, power-consuming traffic lights and replace with roundabouts, which extensive research confirms are many times more effective and self-regulating of traffic flows, and safer, cheaper to install and maintain, more attractive in the environment, produce far less emissions - both in installation and operation, etc, etc, etc...
Tom Brown
Carnegie Mellon University is deploying Its tech called Surtrac at a few placed. AI to monitor what I understand to be several intersections at once to help traffic flow. Significantly fewer emissions when the vehicles flow through intersections rather than stop and idle then go. We can’t automatically make all these existing SUV’s and pick up trucks and tractor trailers have improved gas mileage but at least letting them not stop and go unnecessarily would help. Plus it’s good for a city’s collective psyche to not sit so much in soul crushing traffic. Deploy these techs way sooner than later please. America isn’t going to convert to roundabouts anytime soon.
windykites
Ben, roundabouts are commonly used in the UK, however they are now often fitted with traffic lights, because the traffic flow is too great. A main road often has a steady flow of vehicles which stop other cars from entering.
Bodger
I will repeat Bodger's Dictum: "Artificial intelligence can never overcome natural stupidity". Make the traffic devices as smart as you like there will always be some jerk texting or playing Candy Crush on his/her phone while trying to drive and that will be enough to screw up any planning the machines might attempt.
BlueOak
We’ve had camera-based smart traffic lights for at least a decade in a large portion of our county in SE Michigan. They generally do work, but not flawlessly. It sometimes feels like their programmers have set limits/boundaries to how reactive to traffic conditions are. Other times they seem very responsive to traffic. And once you “get to know an intersection”, their active response is very noticeable.
Nelson Hyde Chick
Nothing infuriates me more than having to wait-out a red light and there are no cars, motorcycles or pedestrians in need of going through the intersection!!!!!!!!!!!!!!!!!!!!!!!
unclejoe
On streets that have the traffic signals
sequenced to the speed limit,
you can have a clear non stop cruise.