Intelligent Detection for Tunnel Shotcrete Spray Using Deep Learning and LiDAR

Abstract

Shotcrete spray is an indispensable process in tunnel construction. At present, the construction of tunnels in China is mainly depend on labor or mobile concrete sprayer, which has lots problems like timeconsuming, low precision, and labor intensive. An intelligent detection method for tunnel shotcrete spraying is proposed in this article. There are two main issues need to be solved, one is the modeling of tunnel in real-time to monitor the thickness of shotcrete and other is the detection of spraying area in the tunnel. The LiDAR can obtain a 3D model of tunnel after performing necessary preprocess on it in real-time. On the other hand, the spraying areas are usually divided by arches in the tunnel, so we can detect the position of arches to determine the spraying areas. Inspired by the YOLO algorithm, we proposed a novel neural network structure to detect the approximate bounding boxes of the arches and a line-detection algorithm is used to determine the final positions of the spraying area in the image. The size of the weight file of our neural network is only 2.57 MB after the use of some deep compression tricks, which means our model is device friendly. After that, the object detection results in the image will be projected to the point cloud data. The experimental results suggest that our method performed well in the detection for tunnel shotcrete spraying, and the mAP for spraying area detection was found to be 91.4%. INDEX TERMS 3D point cloud, deep learning, object detection, tunnel shotcrete.

Application of Terrestrial Laser Scanning (TLS) in the Architecture, Engineering and Construction (AEC) Industry

“…Due to the great advantage of transforming big data into useful knowledge, there is no doubt that AI in collaboration with TLS will be one of the primary future trends in the field of AEC. In recent years, various AI techniques, deep learning in particular, have been found in previous studies to have huge potential in object detection and quality assessment [ 90 , 271 , 272 , 273 , 274 ]. Potential AI-based solutions in future may include: (a) prediction of project activities (e.g., safety, progress, and productivity); and (b) decision-making optimization (e.g., project planning, scan planning, and resource management).…”

Applications of Terrestrial Laser Scanning (Tls) in Mining: A Review

“…Today, terrestrial laser scanning technology is widely used in numerous mining engineering applications due to its outstanding advantages. These applications cover the following subjects:  Determination of discontinuities in open pit and underground mines (Slob et al2004;Slob et al, 2005;Kemeny et al, 2006;Decker, 2008;Sturzenegger and Stead, 2009;Maerz et al, 2012;Lato et al, 2013;Feke¬ te andDiederichs, 2013, Deliormanlı et al, 2014),  Calculating shotcrete thickness in tunnels, determining possible deformation measurements in tunnels, automating lighting fixtures in tunnels (Chun-Lei et al, 2019;Puente et al, 2014;Delaloye and Diederichs, 2011;Fekete et al, 2009),  Conducting deformation measurements to control rock falls and landslides (Bauer et al, 2005;Aksoy & Ercanoglu, 2006;Özdoğan et al, 2018;Teza et al, 2008;Abellán et al, 2008;Salvini et al, 2013),  Designing 3D slope and bench geometry (Feng & Röshoff, 2004;Yanalak, 2005;Oparin et al, 2007),  Conducting different volume calculations (Yanalak and Baykal, 2003;Yakar et al, 2008),  Analyzing blasting results (Engin and Maerz., 2016).…”

Sensing Method for Wet Spraying Process of Tunnel Wall Based on the Laser LiDAR in Complex Environment

“…LiDAR-based 3D object detection is essential for automating tunnel shotcrete spraying, as it directly relates to understanding the tunnel environment. A previous study proposed a novel neural network structure based on LiDAR for detecting the area of tunnel shotcrete spraying [21], demonstrating the real-time monitoring of the tunnel profile and shotcrete area. Ranjbarnia [22] studied the effects of various parameters such as sprayed concrete thickness, soil geomechanical properties, tunnel depth, and fault plane dip angle using the 3D finite difference analysis algorithm and centrifuge physical model, but mainly focused on crossing faults in urban tunnels, which is not applicable for the construction analysis of tunnels in progress.…”