Imagine your smart devices could think and react instantly without waiting for a distant cloud server to answer. This project builds a high-speed digital 'brain' located right where the action is, making it faster and more reliable. It also adds a security layer to ensure these AI decisions are safe, private, and easy to understand.
Current cloud systems are too slow for high-demand apps like autonomous driving or industrial XR. Additionally, AI models at the edge are often 'black boxes' that are vulnerable to attacks and lack privacy.
A modular edge platform for distributed AI and a set of security tools for model verification. These were validated through 7 real-world demonstrations.
If you are a factory operator dealing with lag in augmented reality (XR) tools on the assembly line — this project developed an edge-cloud continuum that provides ultra-high computational performance. This ensures that digital overlays and IoT data appear in real-time without stuttering.
If you are a transit authority dealing with the safety of driverless vehicles — this project developed an edge-assisted autonomous tram system. It uses AI to manage network resources for extremely low latencies, ensuring the tram reacts instantly to obstacles.
If you are an AI vendor dealing with adversarial attacks or data leaks in distributed systems — this project developed a suite of tools for safety verification and explainability. This makes AI model decisions transparent and protects them from malicious manipulation.
Based on available project data, specific pricing or cost structures are not provided as this is a research and innovation action.
Yes, the project demonstrates scalability through 7 demonstrations across two real-world use cases, including industrial sites in Turkey and tram operations in Florence.
The project focuses on open and secure architectures and intends to disseminate results through open sources and standardization bodies.
It integrates via a multi-access RAN and a multi-domain edge-cloud continuum to support B5G (Beyond 5G) applications.
The project period runs from 2023-01-01 to 2025-06-30, with results progressing through various TRLs during this window.
The consortium is heavily weighted toward commercial application, with 11 industry partners (65% ratio) and 3 SMEs. This strong industrial presence, spanning 8 countries, suggests the technology is being developed with direct market requirements in mind rather than purely academic interest.
Contact Universitat Politècnica de Catalunya
Contact us to connect with the VERGE consortium for licensing and pilot integration.
