69 / 2023-03-31 17:54:00

Accuracy and Applicability Evaluation of an Underground Geomonitoring Robot System Using SLAM Methods

geomonitoring,robot,SLAM methods,accuracy and precision,applicability

Geomonitoring provides quantitative and reliable information that is crucial for studying and understanding the processes in the natural and the man-made environment. It helps to identify hazards and to timely adopt appropriate measures. Thus, geomonitoring is one of the most important parts of the entire mining value chain. However, doing this job inherently exposes monitoring staff to a dangerous environment, especially in the field of underground geomonitoring where the working environment is extremely harsh, and adverse ground behavior events such as convergence and ground falls occur frequently. Today, autonomous mobile robotics show promise in the context of "Intelligent Surveying and Mapping" as robots are widely used in various fields. Since 2000, many studies have focused on the establishment of automated robotic systems as well as underground navigation and mapping. However, only a few studies have conducted quantitative evaluations of the proposed or used methods, and almost none have provided systematic and comprehensive assessment of suitability of mapping robot for underground geomonitoring areas. This gap in research motivated the present study of the Institute for Mine Surveying and Geodesy at Freiberg University of Mining and Technology (Technische Universität Bergakademie Freiberg). In this study, the accuracy and precision of the Simultaneous Localization and Mapping (SLAM) methods, implemented on the designed robot system, were systematically and quantitatively evaluated using mine surveying methods. The applicability of SLAM technology for mine surveys and geomonitoring is also assessed. As the subject of this assessment study, a multipurpose robotic research system developed for autonomous geomonitoring and surveying and mapping of underground mining areas was brought into a real underground mine environment 150 m down at the silver mine “Reiche Zeche” in Freiberg, Germany. This wheeled mobile robot is equipped with many modern and high precision sensors such as an inertial measurement unit (IMU), 2D and 3D LiDAR scanners, a depth camera, a hyperspectral camera system and radars. In order to assess accuracy, the design of a permanent underground test site, including the configuration of the control points, selection of targets and optimization of the design was conducted. Measurement experiments utilized a 3D LiDAR, a depth RGB-D camera and IMUs mounted on the robot with open-source SLAM methods, were performed at this test site. The obtained result point cloud was compared with the reference point clouds measured by a total station and a terrestrial laser scanner (TLS). The accuracy and precision of the selected SLAM methods as well as the verifiability and reliability of the results were evaluated using the German Ordinance on the Survey Work and Observations of the Surface Markscheider Bergverordnung (MarkschBergV) as a regulation. Finally, the focus of the discussion is on what extent camera-based and LiDAR-based SLAM methods are suitable for underground geomonitoring, which method is better suited, what factors affect the accuracy, and whether data collected by different sensors can be fused to improve measurement accuracy.