Underground » Geology
This project evaluated an off-the-shelf mobile laser scanner (ZebRevo) for automated mapping of structural discontinuities and other geotechnical features in underground coal mines.
The data collected in handheld mode incurred low mapping drift over a medium scale (≈ 500 m) and provided sufficient data resolution for structure mapping, change detection and feature identification. In contrast, the vehicle mounted mode was only found suitable for geometry documentation, had low resolution and suffered from mapping drift over a large scale (> 2 km) due to inertial sensor bias and feature complexity of mines.
To establish a spatial reference framework in global navigation satellite system (GNSS) signal denied underground mines, the project developed three-dimensional unique identifier (3DUID) tags. The tags are inexpensive to construct, easy to install/recognise and aid in automated georeferencing and coregistration of point clouds for geotechnical applications. Point clouds of scan length up to 1 km could be automatically aligned with errors of < 0.3 m using sets of reference 3DUIDs. The observed accuracy is satisfactory for applications such as change detection, deformation monitoring and structure mapping.
Laser scanning has proven to be an efficient technology for collecting rock mass point cloud scans to facilitate digital structure mapping. However, the development of computationally efficient and robust analytical workflows to map structures in underground mines remains an open research problem. The project developed a new approach, termed the amplitude and phase decomposition (APD) approach, to characterise discontinuities by capturing unique signatures in the form of a sinusoidal wave. Moreover, APD removes the presence of artificial features and non-discontinuity points to reduce uncertainty in the results while requiring minimal computation time. Validation of results against the ground truth indicates that APD can capture discontinuity orientation and dispersion with errors of less than 3°s. The APD approach is fully automated with adaptive parameters for universal applicability in both underground as well as surface operations. The project also developed an automated approach to classify roof bolts in an underground mine for roadway support compliance and monitoring with nearly 81% accuracy in a low-resolution point cloud.
The report identifies that low resolution scanner, such as ZebRevo, when used strategically in handheld mode provides sufficient data resolution for structure mapping and roadway monitoring. However, the robustness of the results is often impacted by the low pulse rate of the equipment and variable point density. A laser scanner with a high pulse rate can substantially improve the mapping outcomes and geotechnical assessment results. The project also identifies the potential for currently implemented laser scanning technology to provide real-time solutions, but the inadequate infrastructure in underground mines for data transfer, geodetic networking and processing capacity remain limiting factors. Nevertheless, laser scanners are becoming an integral part of mine automation due to their affordability, accuracy, and mobility.