Human-in-the-Loop Robot Learning for Smart Manufacturing: A Human-Centric Perspective

Abstract

Robot learning has attracted an ever-increasing attention by automating complex tasks, reducing errors, and increasing production speed and flexibility, which leads to significant advancements in manufacturing intelligence. However, its low training efficiency, limited real-time feedback, and challenges in adapting to untrained scenarios hinder its applications in smart manufacturing. Introducing a human role in the training loop, a practice known as human-in-the-loop (HITL) robot learning, can improve the performance of robots by leveraging human prior knowledge. Nonetheless, the exploration of HITL robot learning within the context of human-centric smart manufacturing remains in its infancy. This study provides a holistic literature review for understanding HITL robot learning within an industrial context from a human-centric perspective. A united structure is presented to encompass different aspects of human intelligence in HITL robot learning, highlighting perception, cognition, behavior, and notably, empathy. Then, the typical applications in manufacturing scenarios are analyzed to expand the research landscape for smart manufacturing. Finally, it introduces the empirical challenges and future directions for HITL robot learning in the next industrial revolution era. Note to Practitioners—This review is motivated by the emergence of the next generation of smart manufacturing, which emphasizes the coexistence of humans and robotics in the manufacturing workstation to mitigate inherent limitations of each. It presents an overview of HITL robot learning-related works to identify state-of-the-art and significant focuses for human-centric smart manufacturing. It classifies representative studies into detailed sub-categories based on various facets of human intelligence, highlighting perception, cognition, behavior, and empathy, providing a complete and detailed survey of this field. The applications in manufacturing scenarios are analyzed, and we discuss the possible challenges and future directions. This paradigm has the potential to revolutionize manufacturing operations, enhancing flexibility, and resilience in supply chains, and efficiency for self-organizing collaborative intelligence and cyber-physical systems toward human-robot coevolution. The goal is to attract scholars in broader research fields to contribute to the development of HITL robot learning for smart manufacturing.