Model-Based Human Robot Collaboration System for Small Batch Assembly with a Virtual Fence

Human-robot collaboration (HRC) systems are being actively applied to manufacturing systems because collaboration between humans and robots in HRC systems provides high flexibility and productivity. However, the application of the HRC system in a small batch assembly is limited by the predefined operational instructions of robots due to frequent changes in customer demands and process plans. In addition, the sharing of workspace between human and robot may occur collisions between human and robot that lead to failure during operation. In this study, we propose a process model-based human-robot collaboration (PM-HRC) system which is able to immediately respond to changes in customer demands and process plans. To do so, we develop an object recognition function to ensure worker safety and a model conversion function to prevent assembly failure by generating and sharing information of assembly process data. A comparative analysis of the PM-HRC system with a commonly used HRC system is performed by using simulation software to evaluate the performance of the proposed system with regard to the programming time and execution time. The results of this study can provide strategic scenarios of using collaborative robots (cobots) for developing a smart factory.