Research On Automatic Cross Section Profiling Theory And Control Strategy For Boom-type Roadheader Based On Machine Vision

The automation and information of fully mechanized mining face have always been the key development direction of the coal mining industry at home and abroad.Aiming at the problems of high risk,poor working condition,high working intensity and low forming accuracy,a positioning and remote control system for cutting head of cantilever road-header based on image and space pose transformation is presented in this paper.This system implements the operator's remote control of the road-header in the video screen,and the automatic forming of the section.Compared with the existing research results of cutting automatically,this method reduces the dependence of the control system on multi-sensor fusion and reduces the cost of the transformation of the road-header.Besides,this method can also be more convenient to combine with coal and rock image recognition algorithm,and realize automatic cutting and cutting of coal and rock in the future,providing more possibilities for the intelligent development of intelligent road-header.The main works achieved in this paper are as follows:(1)Established the motion model of the cutting arm of a longitudinal cantilever type road-header.Calculate kinematic equations and operating envelopes of the cutting head by using the D-H coordinate transformation and the kinematics analysis of the robot.(2)Tracing the cutting head in video by using STC algorithm.Using the combination of feature recognition and coordinate transformation to solve the problem of characteristic fuzziness.(3)The paper discusses the control strategy of cutting head and effect of network delay on control stability and accuracy and use Smith predictor to compensate network time delay and reduce the shake of cutting arm.(4)This chapter provide a video remote control system based on B/S framework,and introduce the composition of server and client part.Besides these,this chapter introduce a method of IMM target tracking,based on particle swarm optimization.At present,the model of the remote control system of the road-header is built successfully.This technology has a wide application prospect in the field of mining and industry.