Cloud-Based Digital Twin for Industrial Robotics

Abstract

Production systems are becoming more flexible and agile to realize the need for more individualized products. Robotics technology can accomplish these demands, but programming and re-configuration of robots are associated with high costs, especially for small- and medium-sized enterprises. The use of digital twins can significantly reduce these costs by providing monitoring and simulation capabilities for the robot and its environment using real-time data. The integration with an ontology as a knowledge base to describe the robot and its 3d-environment enables an automatic configuration of the digital twin and the particular robot. In this paper, this concept is coupled with cloud-computing to enable an effortless integration as service in existing cloud architectures and easy access using the common web-technology-stack for the end-users. A novel architecture is presented and implemented to incorporate the real system with its digital twin, the ontology and a planner to infer the actual operations from the knowledge base. Finally, the implementation is applied to the industrial manufacturing domain to assemble different THT-Devices on a PCB to evaluate the concept.