Study On Path Planning Of Shotcreting Manipulator In Mine Roadway

Shotcrete support is a basic form of mine roadway support.It can not only support the roadway,but also balance the surrounding rock pressure,fill and level the roadway wall.The traditional shotcreting operation mainly uses shotcreting machine for manual operation,which has the problems of poor working environment,high work intensity and low efficiency.In order to improve the automation of shotcreting operation,reduce the labor intensity of workers and improve the working environment of shotcreting operation,this thesis studies the shotcreting path planning of mechanical arm of shotcreting robot system.Based on the information extracted from the threedimensional reconstruction model of the roadway,the independent planning of shotcreting path of shotcreting mechanical arm is realized,which is of great significance to improve the efficiency of roadway shotcreting support and realize automatic shotcreting operation.The main contents of this thesis are as follows:Firstly,based on the principle of roadway design and excavation and the characteristics of shotcrete operation environment,eight types of roadway simulation models are constructed.Based on the automatic shotcreting process and referring to the existing shotcreting machinery,a shotcreting manipulator is designed to replace manual shotcreting.The forward and inverse kinematics,workspace,joint stiffness and anti overturning analysis are carried out to verify that the performance of the shotcreting manipulator meets the requirements.Secondly,in order to improve the intelligent level of shotcreting equipment and realize the autonomous path planning of shotcreting manipulator,this thesis first carries out three-dimensional surface reconstruction based on roadway point cloud.According to the reconstruction results,the roadway size,roadway type and surface morphology are analyzed by using point cloud processing technologies such as cartographic expansion,point cloud segmentation and European clustering.Then,the shotcreting path of the shotcreting manipulator is divided into two stages: flat shotcreting path and filling path.The point cloud analysis results are used as the basis for the selection of path constraints,and the path planning algorithm of the manipulator in the two stages is studied.Finally,in order to verify the correctness of the manipulator analysis and the effectiveness of the algorithm,this thesis builds a simulation environment based on the ROS system architecture,and carries out several simulation shotcreting experiments on the eight types of roadway simulation models,which preliminarily verifies that the proposed roadway spatial feature extraction algorithm can extract the roadway spatial feature information that meets the accuracy requirements;The proposed path planning algorithm can adapt to different shotcreting conditions to plan the shotcreting path,and the shotcreting effect along the path meets the requirements of national standards.Based on the mirobot six axis manipulator,a spray simulation experiment system is built.The Kinect depth camera is used to scan the actual point cloud and compare it with the virtual point cloud to verify the accuracy of the virtual point cloud;Based on the virtual point cloud,the characteristics of the simulated roadway inner wall are extracted and the shotcreting path is planned;Using this spraying path,the wlkata mirobot six axis manipulator is used to simulate the spraying process.By comparing and analyzing the data of the planned theoretical path and the experimental path,the feasibility of path planning and mechanical arm spraying based on virtual point cloud is further verified.