A Digital Twin Implementation Framework for a Collaborative Robot Based on Iso 23247

Digital Twin (DT) technology, also known as a digital replica, is an important technology in Industry 4.0 that allows for the creation of a virtual copy of a physical object. This paper presents a real-time machine health monitoring digital twin (DT) framework for robot machine-tending applications based on ISO 23247. The framework utilizes OPC Unified Architecture (OPC UA) for communication and Robotics Operating Systems (ROS) for data exchange between components. The physical space includes a cobot Universal Robot UR10e, an electronic gripper On-Robot RG2, a depth camera Realsense D435, and an inspection camera Sick PIM50. The cyberspace DT layer consists of three entities: the data collection and device control entity (DCDCE), the core entity (CE), and the user entity (UE). The DCDCE handles data collection and device control via ROS, while the CE performs edge computing for autonomous decision-making using machine learning and artificial neural network models. The UE facilitates cloud-human interaction. The evaluation of the DT framework focuses on comparing the DT's predictions of cobot joint positions and tool center point (TCP) with the real values. The evaluation metrics include the square root of the mean squared error (RMSE), mean absolute percentage error (MAPE), and R2 Score. The results demonstrate that the sensor-driven DT accurately replicates the cobot's tool center point (TCP) and joint positions during simulation, indicating successful twinning between the virtual and physical assets.The proposed framework offers improved process repeatability, part reproducibility, and model interoperability in machine-tending applications.