Cities in a Jam: Reducing Urban Traffic Congestion

In Pittsburgh, a smart traffic lights system has been proven to reduce travel times by 25 per cent, the number of stoppages by 30 per cent and idle time by 40 per cent.

Stephen Smith, a Research Professor in Robotics at Carnegie Mellon University, spotted an opportunity to use his field of interest in the area of traffic. He explained: “You really need technology where multiple decision-makers have to collaborate together to achieve co-ordinated behaviour.”

The system is based on a ’decentralised’ approach. Each intersection controls its own local traffic by watching the approaching traffic through video cameras and even radars, then building a model of that traffic. In real-time, the intersection builds a timing plan for allocating the green light in order to move the traffic in the most efficient way possible. It then sends that plan to the hardware controller that manages the lights and also communicates it to neighbouring intersections.