| Title: |
Artificial Intelligence at the Edge - A joint European roadmap for Edge AI |
| Authors: |
Azzoni, Paolo; Bierzynski, Kay; Daaldero, Gerardo; Dallemagne, Philippe; Diaznava, Mario; Duranton, Marc; Ecker, Wolfgang; Flak, Jacek; Hausrotter, Andreas; Katkoria, Deepak V.; Langer, Jan; Lindgren, Anders; Magno, Michele; Mathis, Harald; Pau, Danilo; Peischl, Bernhard; Perlo, Pietro; Sawyer, Davis; Seifert, Inessa; Solanti, Petri; Taube, Markus; Tedesco, Salvatore; Vögel, Hans-Jörg; Weimer, Lars |
| Publisher Information: |
European Association on Smart Systems Integration |
| Publication Year: |
2025 |
| Collection: |
University College Cork, Ireland: Cork Open Research Archive (CORA) |
| Subject Terms: |
Edge AI; Generative AI (GenAI); EPoSS; INSIDE |
| Description: |
In recent years, digitalisation, the availability of data and the possibilities for applying Artificial Intelligence (AI) have become important business drivers for Europe’s key industrial sectors. In our understanding, AI is a technical system that has the ability to mimic human intelligence, which is characterised by behaviours such as sensing, learning, understanding, decision-making and acting. Due to the availability of powerful computing hardware (graphics processing units (GPUs) and specialised architectures) and large amounts of data, AI solutions – in particular Machine Learning (ML), and more specifically Deep Learning (DL) – have found numerous and widespread applications over the last two decades (including image recognition, fault detection and automated driving functions). Low latency, privacy, connectivity limits and distributed applications have driven research in Edge AI, which enables processing and decision-making near data sources – across cloud, edge, and Internet of Things (IoT) devices. It involves training AI models in the cloud and deploying them on edge devices. In 2021, the EPoSS Edge AI Working Group published a white paper called “AI at the Edge” [1], which provided a broad overview of AI methods and techniques, together with technological milestones to guide the research and innovation over the next few years. Following the publication of this white paper, two industry associations – EPoSS and INSIDE – joined forces. The joint Edge AI Working Group is a community of hardware and software experts from industry and academia who drive research and innovation for both national and EU-funded projects, and contribute their insights and views concerning the future of Edge AI. Recent breakthroughs, and in particular in the domain of Generative AI (GenAI), have driven a clear need to revise our roadmap, including the technology milestones, to better understand and exploit the potential of GenAI in the computing continuum, including at the edge. Figure 1.1 shows how to read our refined and ... |
| Document Type: |
book |
| File Description: |
application/pdf |
| Language: |
English |
| Relation: |
https://www.smart-systems-integration.org/app/uploads/2025/10/EPoSS-INSIDE-RoadmapEdgeAI_Oct-2025.pdf; 63; https://hdl.handle.net/10468/18198 |
| Availability: |
https://hdl.handle.net/10468/18198 |
| Rights: |
© 2025, EPoSS e. V. Permission to reproduce any text for non-commercial purposes is granted, provided that it is credited as source. ; https://creativecommons.org/licenses/by-nc/4.0/ |
| Accession Number: |
edsbas.8DB0B79D |
| Database: |
BASE |