The broad IMOCO4.E challenge is to bridge the gap between the latest research results and best industrial practice in digital twins, AI and advanced mechatronic motion control systems. IMOCO4.E strive to create solid and unimpeachable knowledge for optimizing machines and production lines over their whole lifecycle. Software and Hardware building blocks (BBs), edge-to-cloud distributed and featuring standardized interfaces, will be developed to deliver a complete IMOCO4.E reference framework. These building blocks will embed the latest thinking from the academic community and, moreover, can be enhanced in future with new research results. The project will deliver a flexible, scalable, future-proofed and fully functional product architecture to be exploited in industry in high-performance motion control applications with several overlaps to health, mobility and supply chain management domains.
In IMOCO4.E project, EDI continues to research and develop intelligent industrial robots. Our first challenge is to demonstrate the ability of an adaptable robot to work on real production lines, where the robot must be able to take objects of different types and sizes from a random pile and place them in the appropriate size sockets. The second challenge is to make this robot to be quickly retrainable by anyone to work with previously unseen objects. To this end, the EDI explores reinforcement learning approaches, AI training in simulation environments, and methods for transferring the learned control policies from the simulators to the real environment (sim2real).