As part of the NOUS’ commitment to transparency and open-source contribution under the EU Innovation Radar framework, we are proud to introduce the 5 core innovations confirmed by the European Commission’s Innovation Radar Initiative. The Innovation Radar is a flagship initiative launched by the European Commission to identify high-potential innovations and key innovators within EU-funded research and innovation projects.
1. Auto-Standardizer: Automated Data Translation for European Data Spaces
Lead Beneficiary: AETHON
Category: New Product | Level: Obviously Innovative
Data silos and conflicting formats often hinder seamless cross-sector collaboration. The Auto-Standardizer solves this by using linguistic processing, machine learning, and human-in-the-loop validation to automatically translate raw datasets into recognized European standards. Released as an open-source component, it enables seamless data reuse across critical domains such as Mobility, Energy, and Green Deal data spaces.
2. Federated Cloud-Edge-HPC Architecture & Open-Source Platform
Lead Beneficiary: NET-INTRA
Category: New Product | Level: Innovative
To reduce digital fragmentation, the NOUS Core Platform delivers a modular, policy-driven reference architecture that integrates heterogeneous cloud, edge, and HPC environments into a unified federation. Aligned with the European Interoperability Framework (EIF) and EIRA, it provides layered IaaS, PaaS, and MLaaS services. The accompanying open-source development platform gives organizations full data sovereignty, regulatory compliance, and secure pipeline deployment across distributed infrastructures.
3. HiveMind: Privacy-Preserving Federated Learning Backbone
Lead Beneficiary: ITML
Category: Significantly Improved Product | Level: Obviously Innovative
Developing robust AI models often requires sharing sensitive operational data—a key bottleneck under strict GDPR and data-sovereignty mandates. HiveMind is a privacy-preserving federated learning (FL) framework that allows multiple organizations to collaboratively train shared AI models while keeping raw data strictly local on their local infrastructure. HiveMind abstracts the complex orchestration of FL training rounds, client scheduling, and aggregation across edge nodes, making multi-organization AI training effortless and compliant.
4. Standalone Zero-Trust Cybersecurity for Heterogeneous Infrastructure
Lead Beneficiary: AEGIS
Category: New Product | Level: Obviously Innovative
Modern research and enterprise computing environments can no longer rely on traditional perimeter-based security. This innovation offers a deployment-agnostic, Zero-Trust cybersecurity framework featuring an encrypted darknet-style overlay network paired with identity-centric access controls. Operating independently of NOUS-specific architecture, it uses Single Sign-On (SSO), certificate-based authentication, and policy-driven authorization to secure user and service-to-service communication across any public cloud, HPC facility, or edge network.
5. Distributed Quantum Reservoir Computing for Time-Series Prediction
Lead Beneficiary: NCSR “Demokritos”
Category: Significantly Improved Process | Level: Obviously Innovative
Targeting complex forecasting tasks such as power load prediction, this breakthrough introduces novel Quantum Reservoir Computing (QRC) architectures designed specifically for the Noisy Intermediate-Scale Quantum (NISQ) era. By demonstrating both hybrid and fully quantum, single and multi-node (distributed) reservoir configurations, the framework delivers higher accuracy than classical baselines. Its hardware-agnostic and modular nature enables scaling across multiple QPUs and seamless integration with hybrid HPC-Quantum setups.
Together, these five building blocks pave the way for an open, secure, and sovereign European cloud interface, demonstrating the transformative potential of combining Cloud, Edge, HPC, and Quantum technologies. Stay tuned to the NOUS project portal for upcoming open-source releases, pilot demonstrations, and collaboration opportunities!
