Enhanced Lightweight Virtual Commissioning Technique for Human-Machine Collaboration Utilizing Meta-Digital Twin and AML Modeling

In the context of highly customized industrial applications, human-machine collaboration systems merge the efficiency and precision of collaborative robots with the adaptability and complex decision-making capabilities of humans. This integration adeptly addresses intricate manufacturing tasks that demand high sensitivity and adaptability. However, the virtual Commissioning of human-machine collaborative design schemes prior to the establishment of such systems presents challenges. These include integrating diverse heterogeneous components, the absence of a unified standard for modeling, and the need for dynamic adjustments to design schemes to enhance Commissioning efficiency across multiple engineers. Moreover, in manufacturing environments where cost and efficiency are critical, there is a demand for lightweight collaborative Commissioning tools that can swiftly resolve issues. To address these challenges, this paper leverages AML standardized modeling capabilities to develop a human-machine collaborative digital twin construction methodology. This methodology is based on meta models, meta-actions, and configurable meta processes. Furthermore, it introduces a granular progressive loading algorithm designed for a cloud-edge operational mode, which, when coupled with multi-view classification techniques, fulfills the requirements of lightweight collaboration in platforms such as web browsers/IPAD/Augmented Reality. The approach is ultimately validated through a case study focusing on battery pack wiring harness assembly units within the new energy battery sector.