End Positioning Method of TBM Cutter-changing Robot Based on Binocular Vision

Abstract

The tunnel boring machine (TBM) is widely used in various tunnel construction due to its high excavation efficiency, good safety performance, and low construction cost. TBM in the digging process mainly relies on the cutter on the cutterhead for rock breaking. Under strong interaction with the rock and soil, the rolling cutter is prone to wear and failure. Due to the difficulty of positioning the end of the existing TBM cutter changer robot, the replacement of the failed TBM cutter is still almost entirely dependent on dangerous and inefficient manual operations. To address the above problems, this paper proposes to add binocular vision to the cutter-changing robot system instead of the human eye recognition cutter system and provide visual feedback to improve the accuracy of robot end positioning. A cutter feature recognition and extraction algorithm based on YOLO-SIFT is proposed for a new integrated cutter system. The end positioning system of the TBM cutter-changing robot based on binocular vision was built.