A Human Robot Collaboration Framework for Assembly Tasks in High Mix Manufacturing Applications

Abstract

Abstract Robots are mainly used in mass production applications. Recent advances in collaborative robots make it possible for humans and robots to collaborate on high-mix, low-volume applications. This paper presents a framework for human-robot collaboration in high-mix applications. Tasks that require delicate or complex operations are assigned to humans. Tasks that are straightforward are assigned to robots. Simulation is utilized to design the assembly cell by placing the robots and sensors in the cell. Computer vision is used to localize parts in the cell, and automated motion planning is used to generate robot trajectories to perform assembly tasks. The cell also tracks humans to ensure safe operations. A unique feature of the cell is its ability to detect contingency situations and alert human operators to resolve them. Humans can use a gesture-based interface to communicate with the cell. We use a satellite assembly case study to illustrate the capabilities of the cell and demonstrate the feasibility of performing complex assembly operations using human-robot collaboration.