Helping People Through Space and Time: Assistance as a Perspective on Human-Robot Interaction

Area of Science:

• Robotics
• Human-Computer Interaction
• Artificial Intelligence

Background:

• Assistive robotics has expanded across numerous task domains.
• Comparing diverse assistive strategies is challenging due to domain fragmentation.

Purpose of the Study:

• To unify disparate assistive robotics domains into a general theory of assistance.
• To define assistance from a robot-centric perspective subordinate to user goals.
• To identify and examine key design axes for generalized assistance.

Main Methods:

• Defined a perspective for understanding assistive robots subordinate to user goals.
• Surveyed existing assistive robotics research.
• Identified three key design axes: interaction with others, action space variability, and action timing relative to users.

Main Results:

• Proposed a definition of assistance focusing on robot control and user goals.
• Identified comparable trade-offs across domains for the three design axes.
• Demonstrated how a broader perspective facilitates cross-domain theorizing in assistance.

Conclusions:

• The proposed framework and design axes offer tools for unifying assistive robotics research.
• Examining assistance holistically enables more generalized cross-domain theorizing.
• This approach aids in developing more effective and adaptable assistive robotic systems.