Ontology-based framework for modelling safety in human-robot collaboration systems

Collaborative robots, commonly referred to as cobots, are designed to operate alongside human workers within a shared environment, thereby enabling direct interactions in close proximity to enhance the automation process. This is a core aspect of Industry 5.0. Achieving successful human-robot collaboration (HRC) is a complex and challenging task involving considerable knowledge, information and data. An ontology stores a structured map of the domain knowledge and can describe aspects such as; objects, tasks, skills, experience, behaviours, and actions of the human-robot team. While there is considerable research on ontologies in the HRC domain, safety considerations are still an afterthought. The safety processes and protocols are typically included after these complex systems are realised causing implementation challenges. This paper outlines example HRC safety protocols, distilling these into data constructs and then integrating directly within one complete HRC ontology. We demonstrate how a structured modelling approach, through an ontology, can integrate HRC systems data and safety protocols into a single unified structure that is able to provide reasoning capabilities. This new approach, integrating safety within an HRC ontology, lays the foundations for developing and implementing robust HRCs that have safety at the heart of their systems.