Modern driverless cars usually have an arsenal of artificial intelligently controlled sensors and cameras to help navigate the tarmac while paying close attention to possible dangers in the surroundings.
However, there’s still quite a long way to full automation of road transport. One of the biggest challenges is safety, and one of the biggest factors here is the fact that modern AI isn’t intelligent at all when it comes to interacting with its environment, considering that any single system doesn’t work together with other systems nearby.
“In simple terms, you could say that ten cameras from ten different angles see better than a single camera. Hence, one car with all its sensors will have a much better view of its surroundings if it could also have information from other cars,” says Lukas Esterle, an associate professor at the Department of Electrical and Computer Engineering at Aarhus University.
He is in charge of a new research project aimed at allowing AI and smart systems – be it autonomous cars or something entirely different like smartphones, watches or computers – to work together and collaborate whenever they please sharing relevant information or processing power.
The project named FLOCKD has received a grant of DKK 2,9 million from the Independent Research Fund Denmark.
The focal point is the phrase “whenever they please”, because the project is aiming for total automation – letting the smart devices themselves decide whenever they want to pool their data in real time and collaborate with others.
“Systems should on their own decide when to reap the benefits of working together. That’s the whole idea with the project – to set the AI free to solve problems and work smarter without requiring human intervention,” Esterle says.
He gives another car-related example:
"One example could be at a crossroads. Each car wants to ensure that there are no people walking over the pedestrian crossing. The cars nearby send information to each other along the lines of: ‘I cannot detect any pedestrians on the crossing. Can you confirm this?’ Thus, the individual car can utilize cameras and sensors in several other cars," he says.
The associate professor emphasizes, that by letting AI intelligently cooperate with other systems will result in faster learning, much more intelligent system behavior and synergetic problem-solving – a win win for all parties.
The project launches officially October 1 and will run for 3 years.