Understanding and Resolving Failures in Human-Robot Interaction: Literature Review and Model Development

Area of Science:

• Human-Robot Interaction (HRI)
• Cognitive Engineering
• Human Factors Engineering

Background:

• Robots in unstructured environments frequently fail, yet designs lack user-friendly failure management.
• Untrained users struggle with robot malfunctions, impacting trust and usability.
• Existing research in HRI on robot failures is limited.

Purpose of the Study:

• To explore user perception, resolution, and communication of robot failures.
• To understand how robot failures influence human-robot interaction (HRI).
• To identify strategies for mitigating the negative effects of robot failures.

Main Methods:

• Conducted an in-depth literature review of 52 studies.
• Synthesized insights from Human-Computer Interaction (HCI), cognitive engineering, and psychology.
• Developed the Robot Failure Human Information Processing (RF-HIP) model.

Main Results:

• Identified key stages in how humans process robotic failures: communication, perception/comprehension, and resolution.
• Highlighted gaps in research, particularly concerning interaction failures, human error, and socio-cognitive determinants.
• Found that technical failures receive more attention than interaction failures.

Conclusions:

• The RF-HIP model provides a framework for understanding human responses to robot failures.
• User-centered failure-handling strategies are needed for effective HRI.
• Further research should address interaction failures and the human factors influencing mitigation strategies.