Driverless cars working in cooperation with each other can speed up the flow of traffic by at least 35 per cent, a study has found.
Researchers from the University of Cambridge programmed a fleet of miniature robotic cars to drive around a multi-lane track to observe how the flow of traffic was affected when one of the vehicles stopped.
The cars were able to avoid a pile-up by sending signals to one another when one vehicle stopped in the inner lane, allowing cars in the outer lane to slow down slightly to allow those in closer proximity to the stopped car to pass it quickly without having to significantly slow.
When the cars were not communicating with one another, a queue quickly formed behind the stopped car as the others waited for a gap in the traffic, as would happen in a real-world scenario.
The small cars contained motion sensors and a Raspberry Pi, allowing them to communicate with one another over Wi-Fi.
Introducing a car aggressively driven by a human to the track caused the other cars to move out of its path, demonstrating how the vehicles could interact safely with regular cars.
The study, which has been presented at the International Conference on Robotics and Automation (ICRA) in Montréal, could prove useful in the future for assessing how self-driving cars communicate and their effect on real roads.
“If different automotive manufacturers are all developing their own autonomous cars with their own software, those cars all need to communicate with each other effectively,” said co-author Nicholas Hyldmar, an undergraduate student at Downing College, who designed much of the hardware for the experiment.
Mr Hyldmar worked with fellow undergraduate Michael He, who designed the experiment’s algorithms, as part of their summer research project last year.
“They then adapted a lane-changing algorithm for autonomous cars to work with a fleet of cars. The original algorithm decides when a car should change lanes, based on whether it is safe to do so and whether changing lanes would help the car move through traffic more quickly,” the report read.
“The adapted algorithm allows for cars to be packed more closely when changing lanes and adds a safety constraint to prevent crashes when speeds are low. A second algorithm allowed the cars to detect a projected car in front of it and make space.
“They then tested the fleet in ‘egocentric’ and ‘cooperative’ driving modes, using both normal and aggressive driving behaviours, and observed how the fleet reacted to a stopped car. In the normal mode, cooperative driving improved traffic flow by 35 per cent over egocentric driving, while for aggressive driving, the improvement was 45 per cent.”