Imagine combining a seasoned engineer's rulebook with a modern AI's ability to spot patterns. This project blends those two worlds so that machines don't just make guesses, but can explain why they made a decision. It's like giving a robot a logical brain and a set of safety guardrails so humans can trust them in a busy workshop.
Industrial companies struggle to adopt AI because data-driven models are often 'black boxes' that lack transparency and reliability. This creates a trust gap that prevents AI from being used in critical safety or high-cost environments.
A hybrid AI architecture combining physics-based and data-driven models, a data visualization toolkit for high-dimensional info, and evaluation methodologies for AI resilience and interpretability.
If you are a satellite operator dealing with unexpected equipment failures in orbit — this project developed a hybrid AI approach for anomaly prognosis that detects failures earlier. This ensures higher reliability for assets in space.
If you are a warehouse automation provider dealing with navigation errors in complex environments — this project developed object recognition and autonomous navigation tools. This allows robots to move more safely and efficiently in logistic chains.
If you are an integrator dealing with safety concerns in human-robot shared spaces — this project developed a trustworthy AI system for human-robot collaboration. This empowers staff through safer cooperation and increased decision power.
The project aims to improve shopfloor efficiency by reducing downtime by 30%, which directly lowers operational costs.
Yes, the project tested its advances in real industrial environments and used three representative use cases to demonstrate technical and operational feasibility.
Based on available project data, specific licensing terms are not mentioned, but the project was coordinated by Thales with a consortium of 10 partners.
The project developed tools to assert adherence to ethical and legal regulations, ensuring compliance with societal standards.
It uses a hybrid architecture that integrates data-driven, symbolic, and physics-based components, supported by a data visualization toolkit for monitoring.
The consortium is heavily industry-driven, with 60% of the 10 partners coming from the industrial sector, including 4 SMEs. Led by Thales, the group spans 5 countries (ES, FR, PL, SE, UK), indicating a strong commercial focus and a high likelihood of immediate practical application in the space and robotics markets.
Contact Thales (France) regarding the ULTIMATE project outcomes.
Contact us to connect with the ULTIMATE consortium for licensing or implementation.
