AI-Enabled Traffic Lights May Make Traffic Jams a Thing of the Past

Algorithms to speed up intersections

Key Takeaways

  • AI technologies could make your commute easier and safer. 
  • German researchers say having traffic lights that use AI technology may keep traffic flowing more smoothly. 
  • Some US cities are already using AI systems to control traffic.
The “KI4PED” project focuses on pedestrians rather than vehicles. People are detected and tracked using LiDAR sensor data and AI.

Fraunhofer IOSB-INA

Getting from one place to another could soon get faster thanks to advances in artificial intelligence (AI). 

A new study out of Germany says having traffic lights that use AI technology may keep traffic flowing more smoothly. It's part of a growing number of advances in AI-powered transportation solutions. 

"Everyday users will be able to plan their trips to avoid times or locations of intense traffic," Bilin Aksun-Güvenç, a professor in the Department of Mechanical and Aerospace Engineering at the Ohio State University and IEEE member who studies smart transportation systems, told Lifewire in an email interview. "It will be easier to find a parking spot whose location will be optimized to be as close as possible to the user's final destination."

Get There Smarter

Commuting to and from work can be a nightmare as cars advance slowly in stop-and-go traffic. Researchers at the Fraunhofer Institute in Germany want to change this with their "KI4LSA" project, which uses artificial intelligence to enable smart, predictive light switching. 

Conventional traffic lights use rule-based controls, but this rigid approach does not work for all traffic situations. Also, the induction loop sensors embedded in the road surface provide only a rough impression of the actual traffic situation. 

The researchers at Fraunhofer are working to address these problems by using high-resolution cameras and radar sensors to capture the actual traffic situation more precisely. The researchers claim this allows the number of vehicles waiting at a junction to be determined accurately in real-time. The technology also detects the average speed of the cars and the waiting times. The real-time sensors are combined with artificial intelligence, which replaces the usual rigid control rules. 

"We used a junction in Lemgo, where our testing is carried out, to build a realistic simulation and trained the AI on countless iterations within this model," Arthur Müller, the project's manager, said in the news release. "Prior to running the simulation, we added the traffic volume measured during rush hour into the model, enabling the AI to work with real data. This resulted in an agent trained using deep reinforcement learning: a neural network that represents the light's control." 

AI May Already Be Controlling Your Commute

AI might already control your drive without you even knowing it. 

Cubic's GRIDSMART detection and its adaptive traffic signal control technology, SynchroGreen, are being installed across the US. Gridsmart uses real-time computer vision technology and deep neural net classification to track and discriminate vehicles, bicyclists, and pedestrians as they approach, enter, and exit intersections. 

The company claims the system improves safety for bicyclists and pedestrians and can get vehicles through intersections more efficiently. 

A conceptual image of inner city traffic managed by smart technology.

dowell / Getty Images

"Applications equipped with automatic incident detection or stopped vehicle detection systems are available, as well as advanced applications that integrate live data and feedback from sources such as parking guidance and information systems or weather data," Jeff Price, the vice president of Cubic Transportation Systems, told Lifewire in an email interview. 

The most significant benefit of AI-directed transportation for everyday users will be increased safety, Aksun-Güvenç said. The number of road accidents, especially those involving vulnerable road users (pedestrian, bicyclist, scooterist), will drop significantly due to increased connectivity among all actors on the road and due to the safety mitigation methods that will be available using this shared data and AI methods. 

"These technologies allow visually impaired pedestrians to safely cross streets, for example," she added. 

AI's biggest impact may eventually be for those using autonomous vehicles. The technology can help with collision detection, dynamic routing, and car-following, Stevens Institute of Technology AI assistant professor and transportation expert Yeganeh Hayeri said in an email interview. Systems powered by AI can also help with overspeeding detection, license plate recognition, red-light passing detection. 

"Within a few years, daily commutes will not only be safer and faster but also allow passengers to concentrate on other tasks while in route or even asleep," Chris Hardee, Head of the Software Technology Group at Lumenci, said in an email interview. 

Correction 02/07/2022: Updated Chris Hardee's title in the final paragraph to correctly reflect his current position.

Was this page helpful?