The stress of traffic jams will ease significantly for motorists when fully driverless cars hit the road, according to a new study by the University of Cambridge.
Researchers from the university programmed a small fleet of miniature robotic cars to drive on a multi-lane track and measured how the traffic flow changed when one of the vehicles came to a standstill, similar to if a driver broke down on a dual carriageway.
They found that the advantage of vehicles working together using artificial intelligence is that they will keep moving smoothly instead of bunching up - as a result improving traffic flow by 'at least 35 per cent'.
The study reviewed how motorists behave currently compared to how autonomous vehicles will react to changing situations ahead.
It used a two-lane scenario, with one vehicle stopping in the left hand lane.
 |
| With autonomous vehicles communicating with each other to drive more cooperatively, cars would be able to move more freely |
When cars couldn't see far enough ahead to avoid the stationary vehicle, they quickly stacked up behind it.
As other cars passed in the right lane, those at a standstill struggled to find an appropriate gap due to the incoming speed of traffic and a queue formed.
The settings were the turned up so the vehicles responded more aggressively, simulating reluctant motorists refusing to allow other vehicles into the moving lane.
This caused an even greater jam.
 |
However, when the cars were communicating with each other and driving cooperatively, as soon as one car stopped in the inner lane, it sent a signal to all the other cars.
This notified those in the outer lane to slow down slightly so that cars in the left carriageway were able to quickly pass the halted obstacle without having to stop or slow down significantly.
When the fleet of autonomous cars was programmed to drive cooperatively, overall traffic flow improved by 35 per cent compared with the standard driving test and 45 per cent compared with the aggressive driving test.
|
 |
| No more jams: Students at the University of Cambridge measured the traffic flow benefits of autonomous cars and found they could reduce congestion by 35 per cent |
The findings from the study were presented this week at the International Conference on Robotics and Automation in Montréal, Canada, and used for future research regarding the development of driverless vehicles.
Michael He, the student who designed the algorithms for the cooperative reactions of the autonomous cars, said: 'Autonomous cars could fix a lot of different problems associated with driving in cities, but there needs to be a way for them to work together.'
Fellow author and hardware expert Nicholas Hyldmar, an undergraduate student at Downing College, who designed much of the hardware for the experiment, added: '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'.
While an entire network of driverless vehicles should - in theory - cut journey times, the transition period from human- to self-driving cars is likely to cause plenty of tailbacks.
 |
| Researchers from the university programmed a small fleet of miniature robotic cars to drive on a multi-lane track and measured how the traffic flow changed when one of the vehicles came to a standstill, similar to if a driver broke down on a dual carriageway |
According to research by the Department for Transport two years ago, when one in four cars on British roads are fully autonomous there will be a 'decline in network performance'.
It added that congestion would only improve once driverless vehicles made up between 50 and 75 per cent of all machines on the road.
Doctor Amanda Prorok from Cambridge’s Department of Computer Science and Technology said more studies needed to be conducted into effectiveness of autonomous vehicles to fully understand their impact on the transport network.
'Our design allows for a wide range of practical, low-cost experiments to be carried out on autonomous cars,' she said.
'For autonomous cars to be safely used on real roads, we need to know how they will interact with each other to improve safety and traffic flow.'
The university confirmed it would be pushing ahead with future research looking at more complex driving scenarios, such as additional lanes, intersections and varying vehicle types.
