“I am both here and there” Parallel Control of Multiple Robotic Avatars by Disabled Workers in a Cafe

Domestic RobotsSocial Robot InteractionRobots in Education & HealthcareFamily CaregiversDisability Service Providers

Title of the Paper

“I am both here and there” Parallel Control of Multiple Robotic Avatars by Disabled Workers in a Café

Paper Information

  • Field of Study: Human-Computer Interaction, Remote Work for Disabled Individuals, Robotic Avatars
  • Keywords: Disability, Avatar, Remote Work, Café, Robotics

Research Background and Issues

  • Identified Problems or Challenges:

    • Current research primarily focuses on testing remote avatar systems in laboratory settings, often involving novice users with little experience in these technologies.
    • Limited involvement of disabled individuals and the specific skills they may possess when using avatar robots.
    • Laboratory experiments fail to reflect real-world scenarios and challenges.
    • Existing avatar robot systems typically support a “one-to-one” control model, restricting users from simultaneously controlling multiple avatars to perform tasks.
  • Significance:

    • Enabling disabled individuals to use avatar robots for multitasking in real-world scenarios not only enhances their mobility but also creates more inclusive employment opportunities in the service industry and other fields.
    • Researching parallel control systems for multiple avatars can provide valuable insights for future robotic technologies and remote interaction design.
  • Research Motivation and Related Work:

    • The connection between disabled individuals and technology has a rich history; they have developed unique skills through years of using assistive technologies.
    • Avatar robots have been proven to help disabled individuals overcome physical barriers, enabling participation in education, work, and social activities.
    • New technological frameworks for multi-avatar control (e.g., parallel avatars, re-embodiment) have not yet been tested in real-world scenarios, especially involving disabled individuals.

Solution

  • Method or Solution:

    • Develop a technology called the “parallel avatar control system,” allowing disabled café employees to simultaneously control multiple avatar robots (used for greeting, taking orders, and serving customers).
    • The system is based on peer-to-peer connections and the ability to share avatars.
  • Innovations:

    • First implementation of a parallel avatar system in a real-world environment, enabling disabled employees to control multiple avatar robots simultaneously.
    • Extended the concept of “shared control” for avatar robots beyond single-avatar remote usage.
    • System design focuses not only on functionality but also on optimizing cognitive load and enhancing interaction experiences.
  • Implementation Steps and Key Technologies:

    1. User Interface Design:

      • Design specialized interfaces for different types of avatar robots, considering various input methods used by disabled individuals (e.g., eye tracking, touch, mouse).
      • Introduce simplified movement and interaction commands to reduce cognitive load, such as preset actions (e.g., waving, nodding).
      • Add visualized maps for navigation of larger avatar robots (OriHime-D).
    2. System Architecture:

      • Support simultaneous connection and control of multiple avatars by users, including voice, camera, and microphone functionalities.
      • Design a flexible role-switching mechanism to allow different employees to share avatars based on task requirements.
    3. Employee Training:

      • Conduct multiple trials and practice sessions to train employees in managing multi-avatar tasks proficiently.
      • Encourage employees to develop personalized work styles and enhance service experiences through playful interactions.

Research Outcomes

  • Specific Results:

    • The parallel avatar system significantly improved the work agency of disabled employees, enabling them to establish continuous connections with customers across greeting, ordering, and serving stages.
    • Video analysis and user feedback validated the feasibility of employees successfully controlling multiple avatars and completing tasks.
  • Advantages:

    • Compared to existing “one-to-one” control systems, the parallel avatar system expanded employees’ range of actions and reduced reliance on other staff.
    • Specific designs (e.g., simplified action commands and unified interfaces) effectively alleviated cognitive load.
  • Experiment or Evaluation Results:

    • Customer Feedback:
      • Customers expressed satisfaction with the service quality and efficiency of the parallel avatar system and felt closer connections with employees.
      • Playful interactions (e.g., “conversations” between avatars) brought surprise and enjoyment.
    • Employee Feedback:
      • Disabled employees felt proud of the opportunity to showcase their skills and expressed a desire for further optimization and system expansion.
  • Limitations and Future Directions:

    • Limitations:
      • Operating multiple avatars increased cognitive load, and certain tasks might be affected by attention dispersion.
      • Technical and hardware constraints, such as limitations in avatar robot movements.
      • System deployment remains focused on specific scenarios (e.g., cafés) and has yet to be extended to other fields.
    • Future Directions:
      • Develop more flexible and autonomous avatar robots to support complex scenarios.
      • Expand similar systems to other industries suitable for remote avatar interactions.
      • Explore large-scale avatar-sharing networks to reduce geographical and physical barriers.

Conclusion

This study provides rich empirical data on the first real-world implementation of a multi-avatar robot parallel control system, involving interaction experiences of disabled employees and customers, system design, and improvement suggestions. It offers valuable reference points for the development of robotic technologies and multi-avatar systems in practical applications.

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/papers/chi/96362/2023

AdRecommended

Learn AI Coding at CodeNow

open_in_newOpen DOI Link
DOI: https://doi.org/10.1145/3544548.3581124
At a Glance

Paper Snapshot

fact_check
dataset
Source
CHI
calendar_month
Year
2023
emoji_events
Award
No award tagged
group
Authors
9 authors
sell
Subtopics
Domestic Robots, Social Robot Interaction, Robots in Education & Healthcare
work
Professions
Family Caregivers, Disability Service Providers
article
Content Status
Full text indexed
hub
Related Papers
0 related papers