Close Menu
    Facebook X (Twitter) Instagram
    Sunday, June 15
    Top Stories:
    • Ant International and Ant Digital Pursue Stablecoin Licenses in Hong Kong
    • Unbeatable Deals on Sonos Speakers and Soundbars!
    • Celebrate Dad: Enjoy Up to 50% Off!
    Facebook X (Twitter) Instagram Pinterest Vimeo
    IO Tribune
    • Home
    • AI
    • Tech
      • Gadgets
      • Fashion Tech
    • Crypto
    • Smart Cities
      • IOT
    • Science
      • Space
      • Quantum
    • OPED
    IO Tribune
    Home » Self-Driving Cars Share Road Knowledge Through Digital Word-of-Mouth
    AI

    Self-Driving Cars Share Road Knowledge Through Digital Word-of-Mouth

    Staff ReporterBy Staff ReporterMarch 1, 2025No Comments4 Mins Read
    Share Facebook Twitter Pinterest LinkedIn Tumblr Reddit Telegram Email
    Share
    Facebook Twitter LinkedIn Pinterest Email

    Fast Facts

    1. Innovative Knowledge Sharing: NYU Tandon’s research team developed Cached Decentralized Federated Learning (Cached-DFL), enabling self-driving cars to learn from each other’s experiences about road conditions indirectly while maintaining data privacy.

    2. Efficient Model Exchange: Vehicles can exchange trained AI models directly through high-speed communication when in proximity, allowing knowledge to spread even among cars that seldom meet, enhancing adaptability to diverse road scenarios.

    3. Dynamic Learning System: Cached-DFL allows vehicles to maintain a cache of up to 10 models, updating their AI every 120 seconds while discarding outdated information, ensuring that they prioritize recent, relevant knowledge for efficient learning.

    4. Broad Application Potential: This technology not only improves self-driving cars but can also be extended to other smart mobile agents like drones and robots, promoting decentralized learning and swarm intelligence across various networks.

    Self-Driving Cars Learn to Share Road Knowledge Through Digital Word-of-Mouth

    A team at NYU Tandon School of Engineering has made a breakthrough in self-driving car technology. They developed a way for these vehicles to share insights about road conditions indirectly. This innovation allows each car to learn from its peers, even when they rarely encounter each other on the road.

    On February 27, 2025, researchers presented their findings at the Association for the Advancement of Artificial Intelligence Conference. The study addresses a key challenge in artificial intelligence: enabling vehicles to learn collaboratively while keeping their data private. Traditionally, cars have shared insights only during brief direct interactions, which slows their ability to adapt to changing conditions.

    “Think of it like creating a network of shared experiences for self-driving cars,” said Yong Liu, the research supervisor. Liu, a professor at NYU Tandon, emphasized the importance of this development. “A car that has driven only in Manhattan could learn about Brooklyn’s roads from others, making every vehicle smarter.”

    The researchers coined their approach as Cached Decentralized Federated Learning, or Cached-DFL. Unlike traditional methods, Cached-DFL does not depend on a central server. Instead, vehicles can train AI models locally and share these models directly with others.

    When cars come within 100 meters, they use rapid communication to exchange trained models. This process allows vehicles to share knowledge from previous encounters, spreading information further than just immediate interactions. Each car can store up to 10 external models and refresh its AI every two minutes.

    To keep the data relevant, outdated models are removed based on a staleness threshold. This ensures that cars prioritize the most useful and up-to-date information.

    The research team conducted simulations using Manhattan’s streets. They found that, unlike conventional methods, Cached-DFL allows knowledge to flow through the network, similar to how information spreads in social media. Vehicles can relay knowledge imparted from others, enhancing the overall intelligence of the fleet.

    This innovative mechanism reduces the constraints of traditional learning methods. Vehicles now act as relays, extending learning opportunities across wider areas. This could prove vital in urban settings, where road conditions often vary significantly.

    The study also highlighted several factors affecting learning efficiency, including vehicle speed and cache size. Faster speeds and frequent communications yield better results, while outdated models can impair accuracy. Additionally, a strategy focusing on diverse models from different areas may improve learning outcomes.

    As AI technology moves towards decentralized systems, Cached-DFL offers a secure way for self-driving cars to learn collectively. This approach may also benefit other networks of smart mobile agents, including drones, robots, and satellites.

    The team has made their code publicly available, inviting further exploration in the field. Researchers received support from National Science Foundation grants and various governmental programs. Alongside Liu and lead researcher Xiaoyu Wang, the team included experts from Stony Brook University and the New York Institute of Technology.

    As self-driving technology continues to evolve, innovations like Cached-DFL could enhance vehicle performance and safety, paving the way for a smarter driving future.

    Discover More Technology Insights

    Learn how the Internet of Things (IoT) is transforming everyday life.

    Access comprehensive resources on technology by visiting Wikipedia.

    SciV1

    AI Artificial Intelligence LLM VT1
    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleChina’s Car Makers Defy Government Call to End Price Wars
    Next Article Chainalysis Report: Crypto Crime Grows More Sophisticated
    Avatar photo
    Staff Reporter
    • Website

    John Marcelli is a staff writer for IO Tribune, with a passion for exploring and writing about the ever-evolving world of technology. From emerging trends to in-depth reviews of the latest gadgets, John stays at the forefront of innovation, delivering engaging content that informs and inspires readers. When he's not writing, he enjoys experimenting with new tech tools and diving into the digital landscape.

    Related Posts

    Gadgets

    iOS 26 Finally Brings Customizable Snooze Times!

    June 15, 2025
    Crypto

    Stablecoins: The New Powerhouse in Crypto – Coinbase

    June 15, 2025
    Space

    Unleashing Power: The Dynamics of Booster Separation

    June 15, 2025
    Add A Comment

    Comments are closed.

    Must Read

    iOS 26 Finally Brings Customizable Snooze Times!

    June 15, 2025

    Stablecoins: The New Powerhouse in Crypto – Coinbase

    June 15, 2025

    Unleashing Power: The Dynamics of Booster Separation

    June 15, 2025

    Beyond Tech: The Human Touch in Carbon Solutions

    June 15, 2025

    Rethinking Gravity: Is It Just Rising Entropy?

    June 15, 2025
    Categories
    • AI
    • Crypto
    • Fashion Tech
    • Gadgets
    • IOT
    • OPED
    • Quantum
    • Science
    • Smart Cities
    • Space
    • Tech
    • Technology
    Most Popular

    Dusk and Cordial Launch Zero-Trust Custody for RWA Assets at NPEX

    February 27, 2025

    Conquer Your Public Speaking Fears with Free VR Tool!

    March 15, 2025

    Samsung Now Brief: A Game-Changer for Hydro Homies!

    June 9, 2025
    Our Picks

    Dusk and Cordial Launch Zero-Trust Custody for RWA Assets at NPEX

    February 27, 2025

    Conquer Your Public Speaking Fears with Free VR Tool!

    March 15, 2025

    Samsung Now Brief: A Game-Changer for Hydro Homies!

    June 9, 2025
    Categories
    • AI
    • Crypto
    • Fashion Tech
    • Gadgets
    • IOT
    • OPED
    • Quantum
    • Science
    • Smart Cities
    • Space
    • Tech
    • Technology
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    • About Us
    • Contact us
    Copyright © 2025 Iotribune.comAll Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.