Abstract: To address the problem that the Floating Connection Mechanism (FCM) of a hydraulic support cannot directly obtain the pose of its components and that the accuracy of the overall pose acquisition is limited, this study proposed a pose monitoring method for the hydraulic support FCM based on virtual-real integration. By using a camera installed on the top beam of the hydraulic support to collect depth and RGB images of the FCM, the system generated three-dimensional coordinate information of the real point cloud and stored it in a specified file. After file conversion, Unity read the coordinate data and created corresponding cubes to map the real point cloud. Digital twins of the hydraulic support, FCM, and scraper conveyor were constructed. Parent-child relationships, physics engines, and colliders were added in Unity. A virtual camera was created to capture the virtual point cloud of the upper surface of the FCM push rod. The Iterative Closest Point (ICP) algorithm was used to register the virtual point cloud of the push rod’s upper surface with the cropped real point cloud to obtain the translation and rotation matrix of the push rod. This matrix was then converted into Unity to drive the movement of the push rod. Under the constraints of the parent-child relationships, physics engines, and colliders, it further drove the movement of other FCM components, thus achieving pose monitoring of all FCM components. Experimental results showed that the proposed method achieved a monitoring error of less than 3.5 cm in FCM displacement, less than 1.1° in push rod pitch angle, less than 0.5° in push rod deflection angle, and less than 0.8° in connector deflection angle, indicating high accuracy in overall FCM pose monitoring.