Imagine if your gadgets could think and make decisions instantly without needing to send data to a giant computer far away. This work builds the 'brains' and wiring needed to put smart AI directly into machines and sensors. It's like giving a tool its own instant intuition instead of waiting for instructions from a home office.
Companies rely on centralized cloud servers for AI, which causes delays, high energy costs, and security risks. There is a lack of efficient local hardware and software to run AI directly on devices.
A set of AI-based electronic components, processing architectures, and middleware. This includes a conceptual framework for requirements and 20 industrial demonstrators.
If you are a smart farming equipment manufacturer dealing with slow data processing in remote fields — this project developed edge processing platforms that enable real-time AI decisions on-site. This removes the need for constant cloud connectivity to manage crops.
If you are an electric vehicle system provider dealing with high energy consumption of AI chips — this project developed energy-efficient hardware architectures that process data locally. This extends battery life while maintaining fast response times for safety systems.
If you are a factory robotics integrator dealing with latency in assembly line quality control — this project developed AI middleware and electronic components that process visual data at the edge. This allows for instant defect detection without network delays.
Based on available project data, specific pricing or cost structures are not provided as this is a research and innovation initiative.
Yes, the project is designed for industrial scale, demonstrating applicability across 20 demonstrators in five different industrial value chains.
Based on available project data, the specific licensing terms are not mentioned, though it involves a consortium of 50 partners including 33 industry members.
The project specifically develops edge AI as a viable alternative deployment option to legacy centralised solutions to unlock ubiquitous AI deployment.
The project period runs from 2022-12-01 to 2026-06-30, indicating that final results and validated demonstrators will be available by mid-2026.
The project is heavily industry-driven, with a 66% industry ratio comprising 33 companies, including 14 SMEs. This strong commercial presence, combined with 11 universities and 5 research centers across 11 countries, suggests a high focus on practical application and market readiness rather than pure theoretical research.
Contact SINTEF AS in Norway for technical specifications on edge processing architectures.
Contact us to identify which of the 20 demonstrators aligns with your business vertical.
