Computer Vision Based Pose Bias Detection of Shield Tunneling Machine

摘要

Shield tunneling machine is a kind of special engineering machinery applied to tunnel excavation. In the process of construction, shield tunneling machine maybe produce azimuth deviation, which influences the direction of tunnel excavation. Artificial laser spot observation method is commonly used in shield deviation correction of engineering. However, this method will increase the burden of operators and is prone to misjudge the deviation. For the practical application, this paper put forward a deviation detection method of shield posture based on computer vision. In this method, the fixed laser light source produces light spot on the chessboard target which is installed in the rear of shield machine. The image of chessboard is extracted through the camera. The markers and laser spot in the image is detected by using image processing and analysis method. Which can reflect the deviation in the process of advance of shield machine. Experimental results have verified the validity of this method.