Artificial Intelligence Advancements for Digitising Industry
Chapter
Published version
Permanent lenke
https://hdl.handle.net/11250/3086683Utgivelsesdato
2022Metadata
Vis full innførselSamlinger
- Publikasjoner fra CRIStin - SINTEF AS [5661]
- SINTEF Digital [2395]
Originalversjon
Intelligent Edge-Embedded Technologies for Digitising Industry. 2020, chapter 6, 167-193. 10.13052/rp-9788770226103Sammendrag
In the digital transformation era, when flexibility and know-how in manufacturing complex products become a critical competitive advantage, artificial intelligence (AI) is one of the technologies driving the digital transformation of industry and industrial products. These products with high complexity based on multi-dimensional requirements need flexible and adaptive manufacturing lines and novel components, e.g., dedicated CPUs, GPUs, FPGAs, TPUs and neuromorphic architectures that support AI operations at the edge with reliable sensors and specialised AI capabilities.
The change towards AI-driven applications in industrial sectors enables new innovative industrial and manufacturing models. New process management approaches appear and become part of the core competence in the organizations and the network of manufacturing sites.
In this context, bringing AI from the cloud to the edge and promoting the silicon-born AI components by advancing Moore’s law and accelerating edge processing adoption in different industries through reference implementations becomes a priority for digitising industry.
This article gives an overview of the ECSEL AI4DI project that aims to apply at the edge AI-based technologies, methods, algorithms, and integration with Industrial Internet of Things (IIoT) and robotics to enhance industrial processes based on repetitive tasks, focusing on replacing process identification and validation methods with intelligent technologies across automotive, semiconductor, machinery, food and beverage, and transportation industries.