A Statistical Model to Determine Biomechanical Limits for Physically Safe Interactions With Collaborative Robots

Area of Science:

• Robotics
• Human-Robot Interaction
• Biomechanics
• Occupational Safety

Background:

• Collaborative robots (cobots) offer ergonomic and efficiency benefits for manual workstations.
• Current safety standards (ISO/TS 15066) for power and force limiting (PFL) rely on preliminary data.
• Verified biomechanical limits are needed for safe human-cobot interactions, especially concerning impacts and pinching.

Purpose of the Study:

• To experimentally determine new, verified biomechanical limits for human-cobot interactions.
• To establish safe limits for impact and pinching forces from blunt and semi-sharp surfaces.
• To investigate and incorporate gender-specific pain thresholds into safety limit calculations.

Main Methods:

• A human-subject study involving 112 participants across 28 body locations.
• Gradual increase of contact force until slight pain was reported.
• Utilized a regression model to account for gender differences in pain thresholds.

Main Results:

• Experimentally verified biomechanical limits for human-cobot interactions were established.
• Pain thresholds differ significantly between males and females in specific body regions.
• A statistical model incorporating gender as a covariate was developed for personalized limit calculation.

Conclusions:

• The study provides crucial, verified data for enhancing cobot safety standards.
• Gender-specific pain thresholds must be considered for accurate biomechanical limit setting.
• The developed model allows for precise, percentile-based safety limits for diverse workforces.