Apply AI: AI-Driven Robotics for Industry: Enabling System Integration and Adoption (IA) (Partnership in AI, Data and Robotics)

Expected Outcome: The Apply AI Strategy emphasises acceleration pipelines to ensure a smooth transition from research to deployment of AI-powered robotics. Projects under this topic will deliver common frameworks and reusable building blocks that can serve multiple sectors and use cases, reinforcing Europe’s ability to bring AI-driven robotics to scale. Project results are expected to contribute to all of the following expected outcomes: Wider and faster deployment of robotics, bridging the gap between technology providers and end-users. Development and implementation of modular and interoperable integration frameworks and solutions, including standardized protocols for data, training and safety testing, evaluation and validation of robotic solutions in key use cases Improved competitiveness of European industries, notably SMEs via the development of advanced robotics systems, intelligent planning and control systems, user feedback rendering techniques and cutting-edge AI innovations Scope: The project will address the current European gap in system integration capabilities for robotics solutions addressing the various needs of industries. The project will aim at disseminating a deep understanding of state-of-the-art robotics components, including both hardware and software, and expertise in addressing interoperability issues for the upskilling of system integrators. To maximise the impact and adaptability of deployed systems, the approach should consider the most appropriate tools to speed up integration processes and suitable AI design, training and inference methodologies, ensuring scalability, transferability, transparency, robustness, flexibility, and real-world applicability in diverse industrial environments, and should remain adaptable to the latest technological developments. Integration frameworks will promote the use of energy-efficient AI models and hardware ('Green AI'), alongside carbon-aware deployment and operational strategies for robotic system. Where relevant, projects should contribute to open and widely recognised standards to foster interoperability and uptake across the robotics ecosystem. To enhance safety and performance, projects may include high-fidelity simulation environments or digital twins as testbeds for training, validation and verification, with measures to ensure smooth transfer from simulation to real-world deployment. By bridging the gap between technology providers and end-users, these integrators will enable the creation of seamless, reliable and scalable robotics systems that can be easily adopted by industries, especially SMEs, thereby supporting more flexible and efficient production processes. The project is expected to deliver: A deployable, modular integration framework, validated through at least three real-world industrial pilots covering different reference scenarios to demonstrate that the approach can be adapted to varied industrial needs and company sizes, including both SMEs and larger manufacturers. This framework should provide, for example, a common software layer, standard interfaces to connect to existing workflow and legacy system, possibly also to connect various robot components, coordinate multiple robots and link them with additional AI tools and IoT environments, as well as tested configuration templates and clear guidelines to ensure safe and efficient use. An Integration Kit, building on this framework, which offers ready-to-use modules, example configurations and practical tools that help system integrators and companies to set up, test and run AI-enabled robotics solutions more quickly and with reduced technical effort. Where relevant, high-fidelity digital twin testbeds should be linked to each pilot, allowing safe and realistic testing and training before deployment, and supporting a smooth transition from virtual models to actual production lines. Reusable, datasets (compliant with relevant regulation and IP protection) and practical benchmark tasks, made a... Source: EU Funding & Tenders Portal (CC BY 4.0).