Advancing Occupational Exoskeletons: Usability Assessment of a Minimalist Calibration Interface

Abstract: Research on ergonomics for work-related musculoskeletal disorders (MSDs) remains a significant challenge for affected individuals, businesses, and society. As the most costly category of occupational health issues, MSDs affect more than one in three European workers, making them the most common work-related illness across all industries. A common occupational task in various industries is manual materials handling (MMH), which refers to the process of manually moving, lifting, lowering, pushing, pulling, or carrying materials, goods, or products. Nevertheless, it also poses a risk of injuries and MSDs due to different factors, such as the physical strain involved, repetitive motion, poor ergonomics, and environmental factors, such as uneven floors, cluttered workspaces, slippery surfaces, or poor lighting. A promising approach to address MSDs in the workplace is the use of an occupational back-support exoskeleton. This is a wearable technology designed to reduce lumbar spine physical strain during lifting tasks. Research has demonstrated that these wearable devices can decrease back-muscle activity by up to 40%, effectively reducing spinal loading during MMH tasks. According to the actuation principle, an active exoskeleton (with sensors, controller and actuators) can be more versatile in terms of configuration than a passive exoskeleton. A specific characteristic of active exoskeletons is the possibility of modifying the control strategy to provide appropriate assistive forces according to the task. This control strategy is modulated through a human–machine interface (HMI), which is the cornerstone of user interaction and the basis of cognition to modify and adjust parameters in a system. In an active occupational exoskeleton, more functions can be adjusted, such as calibration, user information (weight and height), and control gains.Purpose: This study presents the usability assessment of the motor calibration function in the novel User Command Interface Round (UCI-R), a minimally adaptable setup system for occupational exoskeletons. Calibration is the first step to set up the exoskeleton XoTrunk before starting the MMH task. A user study was conducted with 10 participants by comparing the original user interface with a newly designed, minimised version. The experiment assessed improvements in user experience and efficiency between the two interfaces.Methods: To systematically evaluate user interaction with the interface, we applied the GOMS (Goals, Operators, Methods, and Selection rules) model, which is a well-established cognitive modelling technique in HCI. In this study, the GOMS was used to compare user actions between the two interface versions with a focus on task flow related to motor calibration. The analysis aimed to estimate the cognitive load, execution time, and operational complexity, providing quantitative and qualitative insights into user experience. The goal of this study was exoskeleton calibration; the operators are interactive cards from the exoskeleton calibration section of the User Command Interface (UCI) and the UCI-R visual interface. The methods are the sequence functions to be performed (to achieve motor calibration) in both interfaces, and the calibration action has a specific rule, which is to remain still during the calibration.Results: Ten subjects participated in the assessment showing differences in usability attributes such as efficiency, satisfaction and time on ask favorable for the UCI-R visual interface.Conclusions: Usability attributes such as effectiveness, efficiency, ease of task, time on task, documentation organisation and satisfaction can lead to the acceptance of a new technology as an occupational exoskeleton. A well-designed user interface improves user experience and satisfaction. The minimised UCI-R yielded higher ASQ and SUM results than the standard version of the UCI.

Keywords: Human Machine Interface, Occupational Exoskeletons, User Command Interface

DOI: 10.54941/ahfe1006736

Cite this paper:

​