The objectives of this study are to investigate usability of anthropomorphic sound feedback for service robot malfunction and to suggest appropriate design guidelines based on the results. Two experiments were carried out: one with a single malfunction and the other with repeated malfunctions. Independent variables were type of feedback (anthropomorphic vs. non-anthropomorphic) and type of malfunctions (catastrophic vs. warning). Usability was evaluated using a questionnaire form, protocol analysis, video analysis, and interviews. Several experimental results were identified including better usability of anthropomorphic feedback over non-anthropomorphic feedback and warning malfunction over catastrophic malfunction, respectively, interaction effect between the types of feedback and malfunction, and the effect of anthropomorphic feedback on people's attitudes toward the feedback.
抄録全体を表示