About RoboSAPIENS

Background

The evolution of autonomous robots requires them to adapt to drastic and unpredicted changes in their system structure or environment. RoboSAPIENS focuses on autonomous robotic software adaptations in response to observed environmental changes including human interaction. The project lays foundations for ensuring such adaptations are carried out in an intrinsically safe, trustworthy and efficient manner, reconciling open-ended self-adaptation with trustworthiness by design.

Overarching Goal

RoboSAPIENS will demonstrate safe robotic self-adaptation on four industry-scale use cases: a new learning assignment for an industrial disassembly robot, a structural change in a warehouse robotic swarm, the prolongation of a hull of an autonomous vessel, and the change of human-robotic interaction considering risk and safety. The research results will be validated up to TRL4.

Objectives

• O1: Enable control software robotic open-ended self-adaptation in response to unprecedented system structural and environmental changes.
• O2: Advance safety engineering techniques to assure robotic safety not only before but also during and after adaptation.
• O3: Advance Deep Learning techniques to actively reduce task uncertainty in robotic self-adaptation.
• O4: Assure trustworthiness of systems that use both deep-learning and computational architectures for robotic self-adaptation.

Expected Impact

With improved and intrinsically trustworthy capabilities, robots of the future will exhibit 'next step autonomy' and flexibility to operate in unprecedented conditions. Safe human-robot interaction will be pushed to the next level: warehouse robot swarms will get more agile, autonomous ships will navigate correctly despite changing conditions, collaborative robots will safely track and avoid nearby humans, and disassembly cells may cope with wildly varying product states.