Novel paradigms and approaches, towards AI-powered robots– step change in functionality (AI, data and robotics partnership) (RIA)

For robots to be usefully and efficiently deployed to perform new activities in physical interaction with the real world requires an improvement in and expansion of the range of functionalities robots can deploy.

This needs to take place in sectors where the capabilities of robots can be utilised to progress productivity in critical industries, support European industries essential for sovereignty and in sectors with high impact across Europe such as manufacturing, healthcare, agri-food, construction etc.

In particular the following major areas of functional performance need to be progressed to the next level of performance:

• significant enhancement of navigation capabilities in order to enhance mobility (underwater, on the ground, in the air, in the body, in areas difficult to reach, on rough terrain, in unpredictable environments, in areas including people or other moving agents, etc.), particularly in highly dynamic and complex environments.
• extension of manipulation capabilities to address:
  • large (of the order of metres to 10s of meters in scale), or heavy (of the order of 100kg to multiple 100kg )
  • or small objects of millimetre or centimetre scale, or smaller; ,
  • or of objects that are soft, deformable, articulated, delicate or hazardous objects;

Each of these require significant advances in precision, force, speed, re-planning, physical perception, grasping, manipulation (including bi-manual), etc.), in order to achieve beyond human capability in manipulation and dexterity.

For large scale manipulation applications include but are not limited to manufacturing, assembly, maintenance and installation of large infrastructure; for example wind turbines, energy pylons, pipelines, dwellings, industrial buildings, transport infrastructure etc.)

For small scale manipulation applications include but are not limited to medical and healthcare (human and animal), pharmaceutical and laboratory automation, process industries, materials processing and micro-fabrication and assembly.

• significant enhancement of functional interaction capabilities to deliver efficient, safe and natural interaction with people, objects, with other robots, within complex and dynamic working environments, including the ability to adapt to variation in the working environment and the needs and dynamics of users, objects and structures, etc.).

Making significant next step advances in these functional capabilities will require paradigm shifts in terms of both physical and systems architecture particularly through the removal of silos between disciplines that contribute to robotics functionalities.

Proposals will need to address safety and security aspects at all levels, as well as consider the data life cycle in line with GDPR.

Proposals should aim to address bold and significant challenges to the enhancement of robot functionality and do so by utilising multidisciplinary research activities.

Proposals should address several of the following in the context of improved functional performance relevant to deployment barriers in a high impact sector:

• Robust perception and the integration of sensing into physical structures to enhance motion and perception
• Advanced safe and reliable navigation functionalities, integrating anticipation, re-planning, high-level goal optimisation. Natural human-robot interaction functionality
• Advanced cognitive capabilities, integrating any type of learning (from experience, collaborative intelligence or learning from human knowledge, frugality in terms of data, unsupervised, etc.), modelling, reasoning, introspection, etc.
• Novel design approaches, e.g. soft robotics, under-actuated, miniaturised, modular/reconfigurable robots including those capable of self-reconfiguration, e.g. for guidance/navigation/manipulations in places hard to reach
• Mobile manipulation, natural manipulation of arbitrary objects including soft, fragile or other items complex to handle (e.g. dirty, slippery, deformable)
• Advanced navigation/manipulation in extreme environments, extremely small and precise in the body, autonomous navigation on shifting and uneven surfaces and in transition, for example between water and air or water and land, field robotics in harsh environments, the handling and manipulation of extremely large/heavy objects, etc.

Where relevant, proposals should contribute to making AI and robotics solutions meet the requirements of Trustworthy AI, based on the respect of the ethical principles, the fundamental rights including critical aspects such as robustness, safety, reliability, in line with the European Approach to AI. Ethics principles need to be adopted from early stages of development and design.

Critical to success will be the interaction of End Users in the definition of the problem domains and use cases that act as barriers to long term deployment and uptake across multiple sectors.

Multidisciplinary research activities should address all of the following:

• Proposals should involve appropriate expertise in the necessary relevant disciplines to reach their objectives. SSH is particularly relevant in addressing human aspects related to human-robot interaction, sensible task distribution between humans and robots, agency, control, trust and handling of data collection, to achieve usability, trustworthiness, safety and adoption of the developed solutions.
• It is essential that scientific and technological results are reproducible and re-usable in order to contribute to the advancement of the targeted research area.
• S&T progress should be demonstrated through use-cases with major and broad socio-economic impact.
• Projects should build on or seek collaboration with existing projects and develop synergies with other relevant European, national or regional initiatives, funding programmes.

All proposals are expected to embed mechanisms to assess and demonstrate progress (with qualitative and quantitative KPIs, benchmarking and progress monitoring, as well as illustrative application use-cases demonstrating well-defined potential added value), and share communicable results with the European R&D community, through the AI-on-demand platform or Digital Industrial Platform for Robotics, public community resources, to maximise re-use of results, either by developers, or for uptake, and optimise efficiency of funding; enhancing the European AI, Data and Robotics ecosystem through the sharing of results and best practice.

This topic implements the co-programmed European Partnership on AI, data and robotics.