CPSoSaware · Project

AI Tools That Keep Complex Connected Machines Running Safely and Securely

digitalTestedTRL 5

Imagine a factory floor where dozens of robots, sensors, and human operators all need to work together without crashing into each other or breaking down. Now imagine a self-driving car that needs to know exactly when to hand control back to the human driver. This project built AI-powered software that watches over these complex machine networks in real time — spotting cyber threats, predicting failures before they happen, and using augmented reality to keep human operators aware of what's going on. Think of it as an intelligent air traffic controller, but for interconnected machines instead of planes.

By the numbers

consortium partners

countries represented

industry partners involved

total deliverables produced

validated use cases (automotive and manufacturing)

50%

industry ratio in consortium

The business problem

What needed solving

Factories and vehicles increasingly rely on networks of connected machines that must work together flawlessly — but when one component fails, gets hacked, or behaves unexpectedly, the whole system can go down. Current monitoring tools were built for simple, isolated machines and cannot handle the complexity of dozens of interconnected devices sharing tasks, making split-second safety decisions, and defending against cyber attacks simultaneously.

The solution

What was built

The project delivered AI-powered failure prediction algorithms, runtime cybersecurity monitoring tools for connected industrial systems, augmented reality interfaces for operator situational awareness, simulation and training data generation tools, and a model-based design engine for dependable system configuration — 35 deliverables in total across automotive and manufacturing domains.

Audience

Who needs this

Automotive OEMs and Tier 1 suppliers developing semi-autonomous or connected vehiclesManufacturers deploying collaborative robots on production linesIndustrial cybersecurity firms protecting operational technology networksSystem integrators building connected factory or smart infrastructure solutionsSafety certification consultants for autonomous and semi-autonomous systems

Business applications

Who can put this to work

Automotive

enterprise

Target: Tier 1 automotive suppliers and OEMs developing semi-autonomous vehicles

If you are an automotive company working on connected or semi-autonomous vehicles — this project developed AI tools that monitor when vehicle systems approach their limits and proactively hand driving tasks back to the human driver. The system was evaluated in mixed traffic environments with semi-autonomous connected vehicles, addressing the critical safety gap between full autonomy and driver control.

Manufacturing

mid-size

Target: Manufacturers using collaborative robots on production lines

If you are a manufacturer deploying collaborative robots for inspection and repair — this project built runtime security monitoring and AI-based failure prediction specifically for scenarios where cobots need to pass control back to human operators in critical situations. The tools were validated with 14 consortium partners across 9 countries for industrial inspection and repair use cases.

Cybersecurity for Industrial Systems

any

Target: Industrial cybersecurity providers and system integrators

If you are an OT security company protecting connected industrial equipment — this project developed runtime cyberthreat monitoring tools tailored specifically to cyber-physical systems. With 7 industry partners involved in development, these tools detect threats across distributed networks of connected devices where traditional IT security falls short.

Frequently asked

Quick answers

What would it cost to implement these tools in our facility?

The project did not publish pricing or licensing costs. As a publicly funded EU research project (RIA), the core tools were developed with public funds, which may allow favorable licensing terms. Contact the consortium coordinator to discuss commercial arrangements.

Can this scale to large industrial deployments with hundreds of connected devices?

The system was designed to work in a decentralized way, with devices collaborating and sharing tasks with minimal central intervention. This architecture inherently supports scaling, though validation was done in controlled automotive and manufacturing scenarios with the 14-partner consortium.

Who owns the IP and how can we license these tools?

IP is shared among the 14 consortium partners across 9 countries under the Horizon 2020 grant agreement. The coordinator ATHINA Research Centre in Greece would be the first point of contact for licensing discussions. With 7 industry partners in the consortium, commercial pathways likely exist.

How mature is the cybersecurity monitoring component?

The project delivered both preliminary and final versions of the runtime security monitoring tools, indicating iterative development and testing. These were specifically tailored to cyber-physical systems and validated against the project's automotive and manufacturing use cases.

Does this work with our existing equipment and software?

The system was designed for heterogeneous computing components including processor cores, GPUs, and FPGA fabric. This suggests compatibility with diverse hardware setups. The AI tools autonomously determine how to allocate processes across different device components, which supports integration with mixed equipment environments.

What is the timeline to deploy this in a production environment?

The project ran from 2020 to 2022 and produced 35 deliverables including simulation tools, AR interfaces, and security monitoring tools. Based on available project data, the tools reached validated prototype stage. Moving to full production deployment would likely require additional engineering and customization work.

Does this meet automotive or industrial safety regulations?

The project addressed dependability, fault tolerance, and security for connected systems. The automotive use case specifically dealt with dynamic driving task handover in semi-autonomous vehicles, which is directly relevant to safety standards. However, specific regulatory certification details are not available in the project data.

Consortium

Who built it

The CPSoSaware consortium is well-balanced for technology transfer with 14 partners spanning 9 countries and a 50% industry ratio — meaning half the team comes from companies, not just universities. The 7 industry partners alongside 4 universities and 3 research organizations suggest that real-world applicability was a priority from the start. The geographic spread across Central and Southern Europe, including Germany, Finland, Israel, and Italy, indicates exposure to diverse industrial standards and markets. With 2 SMEs in the mix, there are partners who understand commercialization pressures. The coordinator is ATHINA Research Centre in Greece, a well-known technology research organization.

ATHINA-EREVNITIKO KENTRO KAINOTOMIAS STIS TECHNOLOGIES TIS PLIROFORIAS, TON EPIKOINONION KAI TIS GNOSISCoordinator · EL
CENTRO RICERCHE FIAT SCPAparticipant · IT
CATALINK LIMITEDparticipant · CY
ATOS SPAIN SAparticipant · ES
TAMPEREEN KORKEAKOULUSAATIO SRparticipant · FI
UNIVERSITA DELLA SVIZZERA ITALIANAparticipant · CH
UNIVERSITY OF PELOPONNESEparticipant · EL
EIGHT BELLS LTDparticipant · CY
ATOS IT SOLUTIONS AND SERVICES IBERIA SLthirdparty · ES
IBM ISRAEL - SCIENCE AND TECHNOLOGY LTDparticipant · IL
FUNDACIO PRIVADA I2CAT, INTERNET I INNOVACIO DIGITAL A CATALUNYAparticipant · ES
PANEPISTIMIO PATRONparticipant · EL
PANASONIC AUTOMOTIVE SYSTEMS EUROPE GMBHparticipant · DE

How to reach the team

ATHINA Research Centre (CERTH), Greece — reach out via the project website or CORDIS contact form for licensing and collaboration inquiries.

Order a report on this consortium See ATHINA-EREVNITIKO KENTRO KAINOTOMIAS STIS TECHNOLOGIES TIS PLIROFORIAS, TON EPIKOINONION KAI TIS GNOSIS profile →

Next steps

Talk to the team behind this work.

Want to connect with the CPSoSaware team? SciTransfer can arrange an introduction and help you evaluate if these AI monitoring and safety tools fit your industrial setup.

Order a report on this topic More in Digital & ICT