Fuzzy Rule-Based Section Forming for Mining Robotic Excavation Error Compensation

Abstract

Based on the automatic forming technology for mine roadway sections and the error measurement technology for boom-type roadheaders, this study primarily uses a fuzzy neural network with a proportional-integral-derivative (PID) controller to compensate for pose errors. By modeling and analyzing the body position and section affecting the boundary accuracy of a boom-type roadheader, boundary compensation is performed according to the error between the preset and actual sections. The compensation accuracy is then analyzed, and the proposed automatic section-cutting control program is further modified to achieve a precise section formation of the comprehensive excavation face. The experimental results demonstrate that there is a decrease of 21 % and 28 % in the horizontal and vertical section errors, respectively; thus, the proposed program effectively reduces the roadway section errors and achieves precise cutting. This study promotes the automation, intelligence, and robotization of coal-mine roadway production and provides data and experimental support for the future development of intelligent boring machine technology.

Keywords

• Cantilever roadheader
• fuzzy neural network
• spatial modeling analysis
• Section shaping
• truncation error compensation