Investigating Human-Robot Interaction: New Insights from Accident Reports

Research conducted at the University of Canterbury's Faculty of Engineering has shed new light on the patterns of relationships between robot characteristics, robot-human errors, and physical working environments in human-robot interaction (HRI) scenarios. By analyzing 303 HRI accident reports using network analysis, the study identified seven HRI incident archetypes, including unexpected activation, faulty commands, and sensor and signal communication errors. The findings have significant implications for safety management and highlight the need for targeted interventions.

Key Takeaways:

• The study identified seven HRI incident archetypes, including unexpected activation, faulty commands, and sensor and signal communication errors, which provide a structured tool for investigating and diagnosing incidents.
• The "Archetype 1: unexpected activation" consistently dominated, accounting for over 60% of accidents, and warrants the most attention in future safety management.
• The increasing frequency of "Archetype 4: sensor and signal communication errors" in later stages highlights the growing need for targeted interventions.
• The research offers a useful framework for researchers and practitioners to understand the patterns of relationships between these factors in different HRI scenarios.
• The study provides a comprehensive analysis of HRI accident reports, highlighting the need for improved safety strategies in robot-assisted work environments.
• The findings have significant implications for safety management, particularly in identifying and mitigating the risks associated with human-robot interaction.
• The research team consisted of Brian H. W. Guo, Yonger Zuo, Yang Miang Goh, and Jae-Yong Lim, all affiliated with the University of Canterbury.

Statistics:

• 303 HRI accident reports were analyzed in the study.
• 7 HRI incident archetypes were identified, including unexpected activation, faulty commands, and sensor and signal communication errors.
• Over 60% of accidents were attributed to "Archetype 1: unexpected activation".

Sources:

• NewsRx. New Androids Study Findings Recently Were Reported by Researchers at University of Canterbury [Identifying Human-robot Interaction (Hri) Incident Archetypes: a System and Network Analysis of Accidents]. Journal of Engineering. November 3, 2025; p 1492.