Scholarlink

Recent advances in MR and AI have paved the way for numerous applications of virtual agents. However, a pressing challenge lies in the interaction with such agents, as they lack physical embodiment, thereby being unable to interact with the real world. Accordingly, we explore the use of a wrist-worn snake-shaped robotic to embody a virtual MR agent. Our implementation comprises an MR agent implemented on an HMD, which is connected to the robotic appendage. We implemented two experiences using our system, a game and an object-finding task. To explore the effects of embodiment, we conducted a comparative study focusing on evaluating the effects of robot embodiment in the two scenarios. Results show that using full embodiment is effective in increasing immersion and realism of the MR agent when interacting directly with it, while an embodiment is not necessary in contexts when visual feedback is sufficient. We discuss the results and future work direction in light of our results.

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

Embodying a Mixed-Reality Agent with a Wearable Snake-Shaped Robotic Appendage

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Abdullah Iskandar,
  • Hala Khazer Shebli Aburajouh,
  • Haya Al Abdullah,
  • Roudha Al Yafei,
  • Noor Al Wadaani,
  • Osama Halabi,
  • Mohammed Al-Sada,
  • Tatsuo Nakajima

摘要

Recent advances in MR and AI have paved the way for numerous applications of virtual agents. However, a pressing challenge lies in the interaction with such agents, as they lack physical embodiment, thereby being unable to interact with the real world. Accordingly, we explore the use of a wrist-worn snake-shaped robotic to embody a virtual MR agent. Our implementation comprises an MR agent implemented on an HMD, which is connected to the robotic appendage. We implemented two experiences using our system, a game and an object-finding task. To explore the effects of embodiment, we conducted a comparative study focusing on evaluating the effects of robot embodiment in the two scenarios. Results show that using full embodiment is effective in increasing immersion and realism of the MR agent when interacting directly with it, while an embodiment is not necessary in contexts when visual feedback is sufficient. We discuss the results and future work direction in light of our results.