Trust‐Triggered Robot–Human Handovers Using Kinematic Redundancy for Collaborative Assembly in Flexible Manufacturing

Abstract

For human–robot collaboration in flexible production, a novel trust-based robot-to-human handover motion planning technique is proposed. A computational model of robot trust in humans is developed, and a method for measuring and displaying trust in real-time is provided. Using this technique, the robot's handover configurations and motions are varied based on robot trust in the human coworker through kinematic redundancy to lessen potential impulse forces on the human body through the payload during the handover. To evaluate the collaborative assembly, including the trust-based handover approach, a thorough evaluation scheme is established. The results of the evaluation suggest that considering the robot's trust in humans during the assembly process, as well as adjusting handover configurations and motions based on the robot's trust levels in humans, improves human–robot interaction (HRI) and overall assembly performance. The results show improved effectiveness in HRI and task performance through increasing safety, team fluency, human trust in robot, handover success rate, and assembly efficiency.