A COMPARISON OF GEOLOGIC MAPS CREATED USING DATA COLLECTED WITH A ROBOT-MOUNTED MOBILE LIDAR, A STATIONARY TERRESTRIAL SCANNER, AND TRADITIONAL MANUAL METHODS: CASE STUDY FROM THE EDGAR EXPERIMENTAL MINE, IDAHO SPRINGS, COLORADO

Geological maps of underground excavations contain information on the structure and composition of an area of interest and are used in mining, underground civil construction, and exploration. Structural data describing discontinuities such as bedding surfaces, joint faces, and fault planes are particularly important input parameters in rockmass classification systems, enabling workers to make informed decisions regarding rock engineering applications. Data required to produce these maps are commonly collected using traditional manual methods which can be inaccurate, subject to sampling bias, and expose workers collecting the required data to hazardous conditions. Manual processing of these data can be inaccurate and time consuming, and as such may not provide accurate information in near real time. An automated method to collect and process these structural data would be safer, more accurate, and faster than traditional methods. A comparison of geologic maps created using data collected with a mobile LiDAR instrument, a terrestrial scanner and traditional manual methods is presented.

Raw data from a case study area in Colorado School of Mines’ Edgar Experimental Mine were gathered using three collection methods:

1. A LiDAR instrument mounted on a mobile robotic platform
2. A stationary terrestrial LiDAR scanner
3. Traditional manual methods

Digital point cloud data collected with the mobile and terrestrial LiDAR were analyzed using manual interpretation and automatic discontinuity set extractor software developed at the Universidad de Alicante, Spain, in order to extract structural data. Data collected using traditional methods were analyzed using manual interpretation in order to identify structural discontinuities.

The orientation of discontinuity planes extracted from the three data sets were plotted on stereonets as poles to planes using RocScience’s Dips software program. Clusters of planes were identified and the number and mean orientation of planes within each cluster were determined. Lastly, the standard deviation of all poles to planes within each cluster was calculated in order to quantify the pole to plane dispersion within each cluster. An analysis of results obtained using three distinct collection methodologies was performed, differences quantified, and results presented.