Algorithm development for automated key block analysis in tunnels from LiDAR point cloud data. (February 2023)

• Record Type: Journal Article
• Authors: Carter-Greaves, L.E.; Eyre, M.; Vogt, D.; Coggan, J.
• Is Part Of: Tunnelling and underground space technology. Volume 132(2023)
• Publication Date: 2023-02
• Subjects: LIDAR -- Kinematic analysis -- Key block analysis -- Scaling and support -- Structural evaluation; Tunneling -- Periodicals; Underground construction -- Periodicals; Tunnels -- Periodicals; Underground areas -- Periodicals; 624.193
• DOI: 10.1016/j.tust.2022.104787
• Languages: English
• ISSNs: 0886-7798

Abstract

Highlight: The algorithm operates in a fully autonomous fashion. Capable of processing point clouds in near real-time, i.e. 9 million points in under one second. Automatically extracts planar surfaces to generate a stereo-net. Uses ISODATA to extract the joint sets. Extends current systems by applying key block theory to identify high risk blocks. Abstract: The process of scaling and support installation in recently blasted tunnels is one of the most hazardous aspects of the underground construction process. An algorithm is developed that can process point clouds captured via LiDAR which can identify the rock mass discontinuities and perform a kinematic key-block analysis. Points clouds containing nine million points can be processed in under one second and the algorithm operates autonomously requiring no human input. The resultant inputs can provide insight into the underlying rock mass and its geo-mechanical behaviour to an entering scaling crew to aid in risk mitigation.