A new type of artificial intelligence (AI) has been designed to help self-driving cars share useful road data with each other, even if they aren’t connected to the internet. This could make future roads safer and help driverless vehicles make better choices in real time.
The Current Problem with Car Communication
Right now, most autonomous vehicles send their data to central computer servers that analyze and store it. While this system works, it has some issues. It puts a lot of pressure on the servers, and if hackers gain access, a lot of information could be exposed at once. Also, in the current setup, cars need to be close together and allowed to connect before they can swap data.
A Smarter, Safer Way to Share
Researchers have created a new system called Cached Decentralized Federated Learning, or Cached-DFL. Instead of sending data to a central computer, each vehicle stores what it learns while driving inside its own system. When it passes near another car, it can quietly share those lessons without needing any internet or prior setup.
For example, a self-driving car that has faced heavy traffic or spotted a broken traffic light can save that experience. When it meets another car later, it can pass on that knowledge, helping others avoid the same issues—even if they’ve never driven through that area
[…]
Content was cut in order to protect the source.Please visit the source for the rest of the article.