Abstract: With the continuous expansion of unmanned aerial vehicle(UAV) and digital twin technologies in engineering applications, new opportunities are provided to enhance engineering monitoring and management levels. Addressing the serious threats posed by adverse conditions in high-slope projects to engineering safety and the surrounding environment, this paper proposes a dynamic feedback method for slope construction safety based on UAV and digital twin technology, aiming to improve the safety and efficiency of slope engineering. Firstly, the method and steps for UAV to collect virtual multi-view digital images of slope engineering are provided through on-site testing. Slope image data is collected, and the SFM-MVS algorithm is used to construct a 3D digital model of the slope engineering.Next, by utilizing a voxel-based spatial geometric analysis algorithm, the excavation width and depth of the slope are dynamically calculated.Then, a point cloud slicing subdivision model is employed, and a structural surface orientation recognition algorithm that considers the spatial extension of the slope outcrop is used to obtain geometric parameters of the structural surface. Finally these data are mapped onto the digital model to establish a digital twin model corresponding to the actual slope.Through continuous data collection, analysis, and simulation, immediate assessment and feedback on slope stability changes caused by construction activities are conducted. The results show that this research method can enhance the timeliness and accuracy of slope stability analysis, providing new prevention and response mechanisms for slope engineering.

Key words: slope engineering, UAV, digital twin, stability analysis, dynamic update