Usability assessment of AI-enabled robots through convergent validity testing of a new questionnaire
摘要
This paper presents the convergent validation of a usability questionnaire tailored specifically to robotic systems, based on the ISO 9241-110 standard and an expansion of applicability for AI-based robotic systems. In prior research, Heinold et al. created a usability questionnaire for robotic systems based on the interaction principles formulated in the ISO 9241-110. The interaction principles are ergonomic design guidelines in human–machine interaction and applicable to human–robot interaction. In a laboratory study we assessed convergent validity of the questionnairs and expanded the validation process to AI-based robotic systems. During the validation study participants interacted with one of two robot types, in one condition AI-based in the other not-AI-based. Cronbach’s alpha was calculated for all factors and compared between the test groups, showing largely good factor consistencies. Convergent validity of the new questionnaire could be largely confirmed with other established usability tools. Group comparisons revealed that the questionnaire is sensetive enough to detect differences in usability between the AI-based and not-AI-based robotics systems. The results demonstrate that the questionnaire performs consistently across diverse robotic systems and is equally applicable to both AI-based and non-AI-based robots, making it a unique instrument within the usability assessment landscape.