Semiautomatic Characterization of Rock Masses Using Photogrammetry, 3D Printing Technology and Ground Penetrating Radar

Rock mass characterization is a way to classify rock mass for scientific, ore prospecting or construction purposes. Currently (2013) rock mass characterization is mainly done using traditional empirical methods in Finnish tunnels and mines. More advanced technologies have existed for some time and they have evolved into affordable, robust and easy to use tools. One example of such technology is photogrammetry which uses multiple 2D photographs to calculate a 3D geometry from the rock surface. This geometry can then be used to determine parameters in a safe environment manually, semiautomatic or fully automatically. One of the benefits in the automatic method is removing the subjectivity from the characterization and increasing the reliability of the analysis. Joint roughness and joint surface alteration are important parameters when considering the joint friction and potential risk of falling wedges or rock blocks. The resolution of the photographic technology is not yet sufficient in the sub-millimeter range and form feeler or similar surface replicating apparatus can be used to capture the joint roughness profile. 3D printing technologies enable production of surface replicas which may be used to study the mechanical parameters of the joint or to manufacture a laboratory testing series. Geophysical methods such as ground penetrating radar (GPR) can be used to assist with the rock mass characterization. The GPR is a versatile subsurface imaging method applied in many geological, geophysical and civil engineering applications. The radar emits electromagnetic impulses which reflect from discontinuities such as bedrock surface, rock joints, ground water level or pipes. It is a non-destructive method, so the studied rock mass remains intact. With the GPR, rock joints can be detected and located inside the rock mass. Furthermore, the GPR can be used to characterize the aperture and filling materials of joints as studied lately. PROBLEM: Rock mass characterization is the classification of rock mass for scientific, ore prospecting or construction purposes. Currently, it is mainly done using traditional empirical methods. This is slow, subjective and dangerous, due to long exposure times in freshly made unreinforced underground spaces. In the last years, new methods have been developed to make partial automatization possible. FACILITIES: 20 m beneath Otaniemi the Geoengineering has an Underground Research Tunnel and supporting facilities. The tunnel is used for research projects and for practical exercises and there is a conference niche and a small workshop for maintenance. The Test Tunnel is also used to teach rock mass characterization. Mobile X-Ray Fluorescence Laboratory was tested in the tunnel during spring 2013. XRFL is a part of SOREX (Service oriented automation for efficient rock and ore exploration) TEKES Green Mining program. In 2012, the Department of Signal Processing and Acoustics did a series of acoustical experiments in the test tunnel using a log-sweep locating. Recently, a series of 48 geophones were installed in a parallel tunnel and drilling induced vibrations were recorded to discover drill bit position. PHOTOGRAMMETRY: One problem is how to acquire a 3D point cloud affordably, quickly and safely. Two solutions exist: Traditional Photogrammetry and Laser Scanning Approach. Photogrammetry is slow, requires heavy image processing and expensive hardware, and has limited accuracy. LADIMO (pat. pend.) was invented to combine the benefits of both methods. A known diffraction pattern is projected and two photographs are taken: one with the pattern (geometry) and one clean exposure with flash illumination (texture). LADIMO is cheaper, faster, safer, smaller and easier to use. 3D PRINTING: 3D printing constructs objects layer by layer from bottom up. It enables the easy production of surface replica molds, which may be used to study the mechanical parameters of a joint in non-destructive way. Another use for the 3D printing is the production of scale models, which are used in verification of calculation methods in complex geometry of coupled physics problems. Compared to traditional replication or scale modelling, 3D printing is cheaper, more accurate and faster (despite the slow printing process). It eliminates previous problems arising from complex or intricate geometries. Currently replicas are being casted into 3D printed molds and further research is needed to enable 3D mineral printers for direct replication. Aalto has several 3D printers available for research projects ranging from rapid prototyping machines at FabLab to professional grade 3D printers at ADD. GROUND PENETRATING RADAR can be used in semiautomated rock mass characterization: emitted electromagnetic pulses reflect from discontinuities (bedrock surface, rock joints, ground water level or objects). The travelling time determines the depth of the reflecting surface. GPR is a completely non-destructive imaging method and the studied rock mass remains intact. Rock joints can be detected and located and information about the aperture and filling materials gathered. IMPACT: Photogrammetry, 3D printing and GPR are great examples of tools available in solving the larger problem. Semiautomatic characterization will revolutionize site investigations and shift the balance from manual labor to automated data processing. In time, fully automated site investigation will be possible and operators will mainly do quality assurance and maintenance. Automated methods are faster, more precise and considerably safer. Best Poster Award