Analyzing psychophysical state and cognitive performance in human-robot collaboration for repetitive assembly processes

Abstract

Abstract One of the main paradigms of Industry 5.0 is represented by human-robot collaboration (HRC), which aims to support humans in production processes. However, working entire shifts in close contact with a robotic system may introduce new hazards from a cognitive ergonomics perspective. This paper presents a methodological approach to monitor the evolution of the operator’s psychophysical state noninvasively in shifts of a repetitive assembly process, focusing on stress, mental workload, and fatigue. Through the use of non-invasive biosensors, it is possible to obtain objective information, even in real time, on the operator’s cognitive load and stress in a naturalistic manner (i.e., without interrupting or hindering the process). In the HRC setting, recognition of the operator’s psychophysical state is the first step in supporting his or her well-being and can provide clues to improve collaboration. The proposed method was applied to a case study aimed at comparing shifts performed both manually and with a cobot of a repetitive assembly process. The results showed significant differences in terms of process performance evolution and psychophysical state of the operator. In particular, the presence of the cobot resulted in fewer process failures, stress and cognitive load especially in the first phase of the work shift. The case study analyzed also showed the adequacy of noninvasively collected physiological data in providing important information on the evolution of the operator’s stress, cognitive load, and fatigue.