Imagine your gadgets and servers acting like a smart crowd at a market. Instead of one central boss telling everyone what to do, each device decides its own price for its processing power and data. They bid and bargain in real-time to team up and solve complex tasks quickly and cheaply without wasting energy.
Current IoT-to-Cloud systems rely on fixed, predefined orientations that are inefficient for dynamic environments. This leads to wasted computational resources, high energy consumption, and security vulnerabilities in heterogeneous networks.
A distributed Middleware Framework featuring Game-intelligent Agents, Collaborative Federated Learning, and RISC-V hardware security acceleration.
If you are a fleet operator dealing with massive data volumes from connected cars — this project developed a middleware that lets cars and edge servers autonomously organize into swarms to process AI tasks. This ensures data sovereignty and efficient resource use across the supply chain.
If you are a factory owner dealing with diverse legacy networks and resource-constrained sensors — this project developed a decentralized system where devices bid to provide computing power. This maximizes service performance while minimizing energy and CO2 waste.
If you are a health provider dealing with sensitive patient data and limited device battery — this project developed secure on-device decision-making and federated learning. This allows medical devices to collaborate on AI tasks without sending all raw data to a central cloud.
The system uses game-theoretic logic where individual devices perform self-assessment to set pricing for their own data and computational assets through bidding and auctioning mechanisms.
Yes, the project aims for a scalable distributed middleware supported by a strong EU-JAPAN ecosystem and validated across 3 vertical and 1 cross-vertical demo.
The project utilizes open-source programming models and APIs to prevent vendor locking and is sustained via the FIWARE Foundation.
It is designed for heterogeneous environments, including legacy enterprise networks and RISC-V hardware for security and AI acceleration.
The project period runs from 2024-06-01 to 2027-05-31, indicating the development and validation phase is currently active.
The consortium is heavily weighted toward industrial application with a 43% industry ratio, comprising 9 industrial partners including 5 SMEs. With 21 partners across 14 countries, including a strategic EU-JAPAN link led by AVL, the project has strong commercial backing and global reach, balancing 3 universities and 6 research centers to ensure scientific rigor.
Contact ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS ANAPTYXIS (REC) in Greece
Contact us to explore integration opportunities with the CoGNETs middleware.
