UAV-MSSH: A novel UAV photogrammetry-based framework for mining surface three-dimensional movement basin monitoring

Abstract

Monitoring the surface three-dimensional movement caused by underground coal mining is a critical and challenging task. Despite many methods have been proposed in previous works, they focus more on the subsidence, with insufficient discussions on the horizontal displacement and the underlying coupled mechanisms between them. To address these issues, we introduce a novel UAV photogrammetry-based framework for mining surface three-dimensional movement basin monitoring, named UAV-MSSH. This framework consists of two branches: vertical subsidence and horizontal displacements. For the subsidence branch, we propose an improved cloth simulation filtering algorithm based on slope zone indexing (CSFS) to model high-precision subsidence basins. In the horizontal displacement branch, we develop a probability integral prediction prior-guided COSI-Corr (PIM-CC) method for displacement detection by bridging the gaps between the horizontal displacement and subsidence. Employing this framework, we monitored the surface three-dimensional movement using the UAV oblique photogrammetry method at a working face in western China. The accuracy evaluation results with net station GNSS data show that the RMSE of the subsidence basin is 161 mm, the relative error is 3.1 %, and the RMSEs of the east–west and north–south components are 99 mm and 85 mm, respectively, which demonstrates the effectiveness and practicality of our proposed UAV-MSSH framework in mining surface monitoring tasks. This study establishes an innovative and proven methodology for UAV monitoring in the mining field, which is valuable for safe production, disaster warning, and ecological restoration.