The SmartEdge project aims to achieve dynamic integration of decentralized edge intelligence while prioritizing reliability, security, privacy, and scalability. This will be realized through a semantic-based interplay of edge devices in a cross-layer toolchain, allowing seamless and real-time distribution of autonomous intelligence swarms. By leveraging low-latency connectivity, the project enables the development of applications that distribute processing, data fusion, and control across diverse sensors, devices, and edges.
To accomplish these objectives, the SmartEdge solution will include a low-code tool programming environment with three main tools:
- Continuous Semantic Integration (CSI) – Ensuring interaction with devices through a standardized semantic interface, employing continuous conversion based on declarative mappings scalable from edge to cloud, and enabling a declarative approach to create and orchestrate apps based on swarm intelligence.
- Dynamic Swarm Network (DSW) – Facilitating automatic discovery and dynamic network swarm formation in near real-time, utilizing hardware-accelerated in-network operations for context-aware swarm networking, and embedding network security.
- Low-code Toolchain for Edge Intelligence – Providing semantic-driven multimodal stream fusion for Edge devices, enabling swarm elasticity via Edge-Cloud Interplay, offering adaptive coordination and optimization, and implementing a cross-layer toolchain for the Device-Edge-Cloud Continuum.
The SmartEdge solution will be extensively demonstrated in four application areas: automotive, city, factory, and health. The project benefits from the collaborative efforts of consortium members including nine industrial partners (Dell, Siemens, Bosch, IMC, Conveqs, Cefriel, Mellanox, IMC.SK and NVIDIA), along with eight research institutes (CNIT, Aalto University, TU Berlin, University of Oxford, Fraunhofer FOKUS, Université de Fribourg, ERCIM) and a standards body (W3C).
Harnessing the power of swarm intelligence in AI
Swarm intelligence consists of a network of endpoint devices that are able to generate and process data at source. An entire network becomes more intelligent and flexible when individual edge devices can identify and share vital information with peers. Sharing this information across healthcare systems and self-driving vehicles must be accomplished while ensuring reliability, security, privacy, and scalability remain uncompromised.