Contemporary Issues and Advances in Human–Robot Collaborations

摘要

Human–robot collaboration is a multidisciplinary research field that was established as an independent research field in the mid-1990s, although several pioneering works predate that time. The first part of this chapter is devoted to giving a brief overview of the history and contemporary issues in human–robot collaborations, while highlighting four key factors: task architecture, team formation, human modeling in the aspect of decision and control, and human modeling of other factors. This chapter presents not only research advances in academia but also technological advances in industry with some real use cases of human–robot collaborations in factories. We then review two of our recent research outcomes, namely (i) human-enabled multirobot navigation, and (ii) operation support with variable autonomy for remote harvesting. In the former, we address a complementary interaction between a human and a network of multiple robots, during which we present a distributed control architecture based on the well-established concept of passivity. In the latter, we address an overlapping human–robot interaction and design an online autonomy determination mechanism based on Gaussian process, a machine learning technique. In both sections, we conduct user studies to examine human passivity, human workload, and human usability. We conclude by providing our vision for the future of human–robot collaborations.