Abstract: Unmanned aerial vehicles (UAVs) have been an important means of monitoring the surface deformation in mining areas since they are flexible, are insensitive to weather, and can swiftly acquire images of target areas. However, there is an urgent need to improve UAV-based measurement accuracy. Based on a systematic analysis of factors affecting the accuracy of surface measurement points, this study focuses on the influence of the quality of ground control points on the point error distribution and accuracy of aerial triangulation. Firstly, through a large number of repeated observations and experiments, this study investigated the error distributions of the planar locations and elevations of monitoring points under the condition of evenly arranged ground control points around. Accordingly, by applying random errors to the planar coordinates and elevations of ground control points, this study determined the corresponding relationship between the planar and elevation accuracy of ground control points and monitoring points. Finally, this study proposes a strategy for improving the UAV-based monitoring accuracy of surface subsidence. The results are as follows: (1) Under the condition of evenly arranged ground control points around, the planar and elevation errors of UAV-based measurement points showed normal distributions but were affected by systematic errors to some extent. (2) The planar coordinates and elevations of the ground control points had independent effects on the planar coordinates and elevations of points in the aerial triangulation. (3) The measurement accuracy can be effectively improved by conducting digital averaging for multiple measurement results. The results of this study will provide theoretical and technical support for the design of UAV-based high-precision monitoring schemes for areas with surface collapse in coal mines.