Collaborative robots, or cobots, have had a significant impact in the industrial environment by facilitating human-robot interaction within shared workspaces. Despite their potential, current implementations often focus on fully automated tasks rather than on true collaboration. This study investigates multimodal bidirectional interactions in a collaborative assembly task, aiming to enhance user satisfaction and usability. We conducted Wizard of Oz (WoZ) tests with 35 participants performing an assembly task assisted by a cobot, evaluating the effectiveness and preferences for different interaction modalities: tactile, vocal, and gesture controls, along with visual, auditory, and haptic feedback. The results indicate that multimodal interfaces are preferred, significantly improving user experience. Visual control and feedback via a graphical tactile tablet were found to be the most effective single modality, while haptic and auditory feedback also played crucial roles in user satisfaction. The study highlights the importance of user-centered design in developing effective human-robot interactions and suggests directions for future research to address the limitations of current experimental setups.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Exploring User Preferences in Multimodal Human-Robot Interaction for the Human-Centered Design of an Assistant Robot

  • Simona D’Attanasio,
  • Augustin Flipo,
  • Juliette Mimault,
  • Anna Studzinska

摘要

Collaborative robots, or cobots, have had a significant impact in the industrial environment by facilitating human-robot interaction within shared workspaces. Despite their potential, current implementations often focus on fully automated tasks rather than on true collaboration. This study investigates multimodal bidirectional interactions in a collaborative assembly task, aiming to enhance user satisfaction and usability. We conducted Wizard of Oz (WoZ) tests with 35 participants performing an assembly task assisted by a cobot, evaluating the effectiveness and preferences for different interaction modalities: tactile, vocal, and gesture controls, along with visual, auditory, and haptic feedback. The results indicate that multimodal interfaces are preferred, significantly improving user experience. Visual control and feedback via a graphical tactile tablet were found to be the most effective single modality, while haptic and auditory feedback also played crucial roles in user satisfaction. The study highlights the importance of user-centered design in developing effective human-robot interactions and suggests directions for future research to address the limitations of current experimental setups.