Key Digital Technologies Joint Undertaking (KDT JU) EdgeAI project kick-off meeting was held in Catania, Italy on 21-23 February 2023, gathering around 70 participants to discuss the priorities in edge AI technology developments and the activities to develop these new technologies over the next three years.

With 49 partners across 11 European countries and a total budget of €35.4 million, EdgeAI is a flagship project for AI technology development at the edge that aims to apply the technologies in 20 demonstrators across 5 industrial value chains: digital industry, energy, agrifood & beverage, mobility, and digital society.

The EdgeAI projects include all the major European semiconductor manufacturers that bring the latest technology developments in edge AI toolchains, new microcontroller architectures and AI-based hardware platforms.

The project addresses the layers of a full-edge AI technology stack and application: sensors, hardware (such as microcontrollers or processing units), SW/HW interfaces, AI frameworks, algorithms, neural network architectures/topologies and new learning data types and methods. In this context, EdgeAI project partners develop technologies that enable the powering of scalable, mission-critical, and industrial edge AI applications.

EdgeAI project aims to positively influence the European Union’s climate-neutral ambitions by having a significant impact on the Green Deal in two directions: one is by increasing energy efficiency and lowering the power consumption of the electronic components, algorithms, and AI methods at the edge and the other is by increasing the processing capabilities and reducing the overall energy consumption and CO2 footprint of the applications that use edge AI components. 

The project aims to substantially impact the European Chips Act as the technology development will identify several challenges that can be put forward as new topics for growth within the next Chips JU European partnership calls.

During three days, the EdgeAI partners presented and discussed the edge AI technological developments and the necessary activities to ensure the adequate planning, implementation and follow-up of the project, including smooth and efficient collaboration between project partners, virtual/physical meetings among partners, preparation of communication materials, networking events, working sessions to exchange practices and to develop results in the 20 demonstrators planned to be implemented. 

The EdgeAI plans to implement activities around the edge AI academy concept by publishing and promoting the project results through the organisation of conferences, sessions, and events aimed at sharing, explaining, and promoting the edge AI technologies and applications developed by the project partners.

Dr Ovidiu Vermesan, Chief Scientist at SINTEF Digital and the coordinator of the EdgeAI project stated during the kick-off meeting: “EdgeAI addresses processing at the edge as a granular continuum encompassing the micro-edge (processing capabilities which are very close to the sensors, microcontrollers, etc.), the deep edge (processing units with extended processing power, such as mobile phones, gateways, etc.) and the meta-edge (micro-servers combining processing units and different microcontrollers for specific operations, for instance). These developments will be partly based on trends observed over the past years that relate to distributed autonomic systems, which employ edge AI techniques to enhance autonomous behaviour in different implementations or collaborative systems.

Edge AI technologies enable more intelligent functionality and can create cross-functional, scalable autonomous systems that have what I call “intrinsic intelligence” and “extrinsic intelligence”. Intrinsic intelligence is the intelligence built within intelligent edge devices, robotic things, or autonomous vehicles, while extrinsic intelligence is the intelligence used for collaboration between these entities in different applications across the industrial sectors.”