Patwardhan, A & Karim, R 2024, 'Ground support condition monitoring through point cloud analytics', in P Andrieux & D Cumming-Potvin (eds), Deep Mining 2024: Proceedings of the 10th International Conference on Deep and High Stress Mining, pp. 631-642, https://doi.org/10.36487/ACG_repo/2465_38
(https://papers.acg.uwa.edu.au/p/2465_38_Patwardhan/)
Abstract: This paper presents a methodology and the results of workflow developed to process point cloud data from underground drifts for condition monitoring of ground support. The workflow focuses on extraction and comparison of information of individual rockbolts and rockbolt neighbourhood prior to, and following, recorded seismic events. Data sources used in this methodology are point cloud data resulting from mobile LiDAR scanning and event data of blasting and microseismic events. In the first step of the workflow, locations in the drift with recorded microseismic events in the vicinity are selected. In the second step, LiDAR scans performed before and after the occurrence of one or more natural or man-made events are used to extract point cloud data within a region close to the recorded events. The extracted point cloud data is processed to compute information about the rockbolts. For each detected rockbolt, the following information is extracted: position on drift surface, tip position, angle to drift surface, length, neighbouring rockbolts, and rockbolt to neighbour’s distances. In the next phase, the rockbolt information extracted from two or more scans over the period encompassing the event are analysed. Corresponding rockbolt information from pre-event and postevent point cloud data are used to compute variation in rockbolt features. The computed variations are examined statistically and used to create a visualisation for decision support to be used by rock mechanics engineers and surveyors.

Keywords: ground support, condition monitoring, point cloud