Safety evaluation of human–robot collaboration for industrial exoskeleton.

• A new industry-oriented physical safety evaluation index system was created for wearable exoskeleton human-robot collaboration using literature analysis, group decision-making, and the Delphi method. • The new index system improved the comprehensiveness of the existing evaluation index system for wearable exoskeleton human-robot collaboration. An industrial exoskeleton combination safety evaluation model was established to address both the physical safety of industrial exoskeleton human-robot collaboration and the perceived safety of human operators. The new model made the industrial exoskeleton evaluation more comprehensive and systematic, and it has the potential to be applied across different human-robot collaboration scenarios. Wearable exoskeletons are a type of collaborative robots that are widely used in human–robot collaboration, and their safety in the collaboration process has received significant attention. An industrial exoskeleton was the research object of this study. First, the influencing factors found in the relevant literature were analyzed and summarized. Subsequently, a group decision-making method was used to formulate preliminarily evaluation indicators. These indicators were then evaluated using the Delphi method. Based on the finalized evaluation indices, a physical safety evaluation index system for industrial exoskeleton human–robot collaboration was constructed, and an evaluation index model was established to evaluate the physical safety. Simultaneously, a questionnaire on the perceived safety was used to evaluate the perceptions of human operators on the safety of industrial exoskeleton human–robot collaboration. These two types of safety, the physical and perceived safety, were combined in an industrial exoskeleton human–robot collaboration safety evaluation. Three different industrial exoskeletons were used in the case study for this research. Their safety was evaluated using the model proposed in this study, which enabled human operators to quickly and effectively select an industrial exoskeleton through comparative analysis. [ABSTRACT FROM AUTHOR]