A Systematic Literature Review of Trust-Aware Shared Control for Human-Robot Collaboration in Construction

Recent years have seen a growing interest in human robot collaboration (HRC), owing to the unique ability to integrate human and machine intelligence collaboratively to complete tasks efficiently. Trust between human and robot is key to the successful deployment and dissemination of HRC. Therefore, trust-aware HRC planning and controlling are now increasingly being considered to ensure that the HRC process is both smooth and comfortable by incorporating real-time human trust. This study takes shared control as the context to examine the state-of-the-art studies of trust-aware autonomy control for HRC. In trust-aware autonomy control, the robot interprets the human trust, initiates the adaptation strategies, and guides the robot to use the appropriate autonomous mode in the next round of its actions. The result revealed that the trust-aware HRC can be categorized as three areas: trust modeling type (qualitative vs quantitative), trust modeling directions (unilateral trust, bilateral trust), and adaptive strategies (autonomous vs teleoperation, haptic force). Based on the finding, a novel HRC trust management framework is proposed consisting of trust inference, trust calibration, adaptive control, and HRC interaction design mode. In addition, the study examined the implications of the framework for a less-digitized industry, in this case, the construction industry.