Abstract: With the increasing demand for flexible production in intelligent manufacture, quadruped robots have become innovative transportation carriers and inspection assistants in industrial applications, owing to their superior mobility and adaptability to terrain. However, quadruped robots are only used as mobile platforms in current practical applications, and their own state informations are not concerned. If the quadruped robot's locomotion system fails, its core operations will be interrupted. At the same time, quadruped robots primarily rely on localized control for locomotion and task execution. When multiple quadruped robots are required to move simultaneously, this way will be time-consuming and labor-intensive. To address this issue, this paper adopts the cloud-network-device framework and applies digital twin technology to integrate the discrete tasks (inspection, transportation) of quadruped robots into the digital twin system, which can observe the current equipment state informations (including quadruped robot) and the implementation of tasks in real time, and has a remote task distribution mechanism to realize the unified and coordinated scheduling of tasks. This research application provides a novel approach for the application of quadruped robots in intelligent manufacturing systems, which is expected to accelerate the evolution of industrial automation towards advanced intelligence. Furthermore, it serves as a reference implementation for the construction of smart factories or intelligent workshops.