This project aims to develop technologies which will enhance the operational capabilities of mobile robots for use in the inspection and maintenance of industrial facilities. A major task in many industrial environments is the retrieval of samples for chemical or biological analysis. Surface swabbing is often conducted manually, but this creates limits on the number of samples that can be taken and their location. There are often many places that can't be reached by people either due to the location (very high, or in confined/restricted access spaces) or environmental hazardous (such as heat or radiation). This project will develop a multi-domain, multi-agent robotic sample retrieval system that will be able to obtain samples across a range of environments. These samples will either be stored for ex-situ analysis in labs or taken to mobile labs for in-situ, real-time analysed. Due to the nature of the operational environments, full autonomy is not desirable, so shared autonomy (human-in-the-loop) will be required. The primary application focus will be nuclear environments, however the technologies will be applicable to many other sectors include petrochemical, offshore and agriculture.

Climate change poses an imminent threat to our civilization. Prominent new technologies to fight climate change involve the earth’s underground renewable and sustainable energy resources and underground storage. However, all these technologies depend on the injection of fluids into the earth’s crust, which, in turn, can cause significant earthquakes. INJECT will solve this problem on the basis of a new, ground-breaking scientific method that will prevent human-induced seismicity and will maximize energy production and storage from renewable and sustainable natural resources. INJECT’s interdisciplinary methodology is based on an astute scientific programme that brings knowledge far beyond the current state of the art. It brings control theory and mathematics to the heart of this new challenging problem. Based on cutting-edge theoretical developments, robust controllers and observers will be designed to optimally adjust fluid injection rates, prevent induced seismic events over large regions and optimize energy production and storage. The controllers will be derived using rigorous mathematical proofs and will take account of the complexity, the heterogeneities and the various uncertainties of the underlying physical processes. INJECT’s innovative theoretical methods will be thoroughly tested through novel numerical models and original experiments. High-fidelity numerical models will account for poro-elasto-dynamics, Coulomb friction, multiphysics and reduced-order modeling, and will outpace any existing algorithms in fault mechanics, both in terms of speed and accuracy. The experimental plan will build on a novel laboratory-scale demonstrator and hybrid lab-computer testing that will be designed and constructed to experimentally validate INJECT’s new concepts. Only then will it be possible to apply INJECT’s methodology in practice and unlock the significant energy potential of the Earth, reduce carbon emissions and help save our civilization.

To develop autonomous robots that are able to comply with social conventions and expectations, and avoid rejection from humans requires that robots must be aware of the social context in which they operate. To this extent, robots need to be endowed with high levels of reactivity, proactivity, responsiveness, and intelligibility. The Doctoral Network - Industrial Doctorates on Social aWareness for sErvicE roboTs (SWEET) aims at training a new generation of research and professional figures able to advance the development of socially aware robots capable of perceiving, interpreting, and responding to human emotions, intentions, and cultural differences. The training program will offer a diverse curriculum, encompassing theoretical knowledge, hands-on technical skills, and real-world application scenarios. The network's interdisciplinary approach includes various fields, such as artificial intelligence, machine learning, human-robot interaction, computer vision, and cognitive sciences. Doctoral candidates will be immersed in cutting-edge research and innovation, gaining insights from the experience of both industrial and academic acclaimed research groups. The network will place a strong emphasis on ethical considerations and responsible innovation, deploying socially aware robots aligned with societal values and promoting inclusivity. Doctoral Candidates will also have the opportunity to participate in a unique coaching program for continuous professional development of their soft and leadership individual skills. Integration Milestones following a Scenario-Based Learning approach will provide co-working activities where collaborative design/implementation is fostered. The inter-sectoral collaboration between academia, user groups’ representatives, business developers, and robot manufacturers of the project will further strengthen the novelty and impact of the research and training, and that research results are economically, socially, and technically feasible.