Researchers have discovered a way for self-driving cars to share information freely while staying on the road without the need to install a direct connection.
“Cash Developed Federed Learning” (Cash-DFL) is one artificial intelligence (A) model sharing structure Self-driving cars This allows them to pass each other and share accurate and recent information. This information includes the latest ways to handle navigation challenges, traffic patterns, road conditions and traffic signals and signals.
Typically, cars are almost next to each other and provide permissions to share the driving insight into the driving insight during their journey. Along with Cashed-DFL, however, scientists have created a semi-social network, where cars can see each other’s profile page of driving discoveries-without sharing the individual information or driving pattern of all the driver.
Self-driving vehicles currently use data stored in a central location, which also increases the possibility of large data violations. Cashd-DFL system enables vehicles to carry data in trained AI models in which they store information about driving conditions and scenarios.
“Think about it such as creating a network of shared experiences for self-driving cars,” written Dr. Yong liuResearch Supervisor and Engineering Professor of Project at Nyu’s Tandon School of Engineering. “A car that is operated only in Manhattan can now learn about the road situation in Brooklyn by other vehicles, even if he never drives there.”
Cars can share how they handle the same scenarios as those in Brooklyn that will appear on the streets in other areas. For example, if Brooklyn has oval -shaped pits, cars can share how to handle oval pits, whether they are in the world.
Scientists upload them Study On 26 August 2024, the preprint RXIV database and on 27 February, they presented their conclusions at the Association for the progress of Artificial Intelligence Conference.
Better self-driving cars key
Through a series of tests, scientists found that quick, frequent communication between self-driving cars improved the efficiency and accuracy of driving data.
Scientists placed 100 virtual self-driving cars in a fake version of Manhattan and set them to “drive” in a semi-yielding pattern. Each car contained 10 AI models that were updated every 120 seconds, which is the cashed part of the experiment. Cars hold on data and wait for it to share until they have a proper vehicle-to-vehicle connection to do so. This is different from traditional self-driving car data-sharing models, which are immediate and do not allow any storage or cashing.
Scientists charted how soon the cars learned and what cashed-DFL made the centralized data system in today’s self-driving cars better than commonly. They found that by the time the cars were within 100 meters (328 ft) of each other, they could see and share each other’s information. There was no need to know each other to share information to vehicles.
“Scalableity is one of the major benefits of decentralized FL,” Dr. Zee JuAssociate Professor in Electrical and Computer Engineering at Florida University told Live Science. “Instead of each car communicating with central server or all other cars, each vehicle only exchanges the model update with those. This localized sharing approach prevents the communication overhead from growing rapidly because more cars participate in the network.”
Researchers imagine cached-DFL to make self-driving technology more inexpensive by reducing the need for computing power, as the processing load is distributed in several vehicles rather than being concentrated in a server.
The next stages for researchers include the real-world test of caced-DFL, removing computer system framework barriers between different brands of self-driving vehicles and enabling communication between vehicles and other connected equipment such as traffic lights, satellites and road signals. It is known as a vehicle-to-some (V2X) standards.
The team aims to run a wide step away from the centralized server and instead is towards smart devices that collect data to collect data and perform the nearest process, which makes data sharing as soon as possible. This rapid herd creates a form of intelligence that is not only for vehicles but for satellites, drones, robots and other emerging forms of connected devices.
“Recentrated federated learning users provide an important approach to learning without compromising privacy,” Javed KhanThe chairman of software and advanced security and user experience at APVI told Live Science. “By locally by cashing models, we reduce dependence on central servers and increase real-time decision making, are important for security-mating applications such as autonomic driving.”
