MIT AI System Helps Autonomous Vehicles Dodge Red Lights
MIT researchers develop a machine-learning approach that enables autonomous vehicles to adjust their speed and avoid idling at intersections. The technology reduces fuel consumption and emissions while keeping traffic moving efficiently.
MIT researchers develop a new machine-learning approach that enables autonomous vehicles to adjust their speed as they approach signalized intersections. By carefully controlling acceleration, the AI system allows self-driving cars to time their arrival so they pass through traffic lights without needing to stop. This technique eliminates the need for vehicles to sit idly at red lights, which currently wastes fuel and generates unnecessary greenhouse gas emissions.
Through extensive simulations, the research team demonstrates that this AI-driven method significantly reduces fuel consumption and emissions while simultaneously improving average vehicle speeds. The system achieves optimal results when every car on the road operates autonomously. However, the researchers find that even if just 25 percent of the vehicles on the road use this control algorithm, the fleet still delivers substantial environmental and efficiency benefits.
The researchers highlight that targeting intersection idling represents a highly practical intervention for combating climate change. Unlike other environmental measures that often require people to sacrifice their quality of life, reducing time spent stuck at traffic lights improves daily commutes while cutting carbon output. As autonomous vehicles become more prevalent, integrating this AI approach could transform urban driving into a smoother, greener experience.