接触に応じて膨張・収縮動作を行うロボットの ふるまいに対する印象評価

Abstract

This study investigates how volumetric body changes in a robot—specifically expansion, contraction, or no morphing—affect users’ impressions during physical contact interactions. We employed a robot MOFU (Morphing Fluffy Unit), a mobile robot capable of full-body morphing, and implemented three types of response behaviors triggered by touch: expansion, contraction, and no morphing (pause only). A within-subject experiment was conducted with 20 participants, and subjective impressions were assessed using the Robot Behavior Anxiety subscale of the Robot Anxiety Scale. The results showed no statistically significant differences in Robot Anxiety Scale scores among the expansion, contraction, and no morphing conditions, as indicated by a Friedman test (χ²(2) = 0.89, p = .64)．However, the Kendall ’s coefficient of concordance with ties correction (W = 0.599) suggested a medium-to-large effect size, and a separate test for W > 0 revealed a significant tendency toward consistent rankings among participants (χ²(16) = 28.75, p = .026)．These results suggest that while the overall differences in anxiety scores were not statistically significant, participants may have perceived and ranked the robot’s morphing behaviors in a consistent manner, providing preliminary insight into how volumetric changes during touch interaction might influence user impressions．