Biomimetic Approach to Designing Trust-Based Robot-to-Human Object Handover in a Collaborative Assembly Task

Area of Science:

• Robotics
• Human-Robot Interaction
• Biomimetics

Background:

• Collaborative assembly tasks require seamless object handover between humans and robots.
• Human trust in robots significantly influences human biomechanics and psychology during handover.
• Existing handover mechanisms often lack adaptability to varying trust levels, impacting interaction quality and safety.

Purpose of the Study:

• To investigate the impact of human trust on human-robot object handover biomechanics and psychology.
• To propose and evaluate a novel biomimetic robot-to-human handover mechanism that adapts to human trust.
• To enhance human-robot interaction (HRI) and overall performance in collaborative assembly.

Main Methods:

• Investigated human psychology and biomechanics (kinetics, kinematics) under varying trust conditions (high, moderate, low).
• Developed a novel robot-to-human handover strategy using kinematic redundancy to minimize impulse forces.
• Implemented and evaluated the proposed mechanism in a human-robot collaborative assembly task within a hybrid cell.

Main Results:

• Decreased human trust led to slower hand movements, increased hand-arm angles, and higher grip forces.
• The novel robot handover mechanism significantly improved HRI metrics: transparency, naturalness, engagement, cooperation, cognitive workload, and human trust.
• Overall performance saw enhancements in handover safety, success rate, and assembly efficiency.

Conclusions:

• Human trust is a critical factor in designing effective human-robot object handover.
• The proposed biomimetic handover strategy enhances safety and efficiency in human-robot collaboration.
• This approach has broad applicability in various industrial and service sectors requiring human-robot cooperation.