Research On Positioning And Navigation Method Of Drilling Robot For Rockburst Prevention Based On Lidar

Impact pressure is one of the most serious dynamic disasters in deep mining,and impact pressure relief by drilling is the main means to reduce impact pressure accidents.At present,the pressure relief operation of drilling holes mainly relies on manual operation,which has great safety risks.Therefore,it is urgent to study the relevant technology of automatic drilling and pressure relief operation of drilling robot for rockburst prevention,and accurate positioning is the premise of automatic drilling of drilling robot for rockburst prevention.Based on lidar SLAM technology,this thesis establishes a roadway environment perception model,solves the problem of positioning mapping and path planning,and provides strong technical support for the accurate positioning and navigation of drilling robot for rockburst prevention,the main work is as follows:(1)The working environment characteristics of the drilling robot for rockburst prevention are analyzed,and the specific requirements for the positioning and navigation system of the drilling robot for rockburst prevention are put forward.Combined with the proposed requirements,the hardware and software system framework of the positioning and navigation system of the drilling robot for rockburst prevention were designed,and the visual interface was developed based on the ROS system design.(2)By analyzing the causes of lidar point cloud distortion problem and the shortcomings of classical lidar point cloud distortion correction model,a lidar point cloud distortion correction scheme based on IMU continuous time trajectory is designed.Aiming at the environmental degradation of roadway,a lidar odometer positioning method based on IMU close coupling is designed based on LOAM algorithm.By introducing SC descriptors and pose maps,a global map construction method based on graph optimization is designed,which reduces the influence of accumulated error on trajectory and map drift.Finally,the proposed algorithm is verified in the coal mine roadway simulation model and public dataset built by Gazebo.(3)The inherent defects of the traditional APF algorithm and RRT algorithm are analyzed,combined with the navigation requirements of drilling robot for rockburst preventions,a local path planning based on improved APF algorithm and a global planning method based on improved RRT algorithm are proposed,and then the path planning and navigation fusion strategy of drilling robot for rockburst prevention is designed,and simulation and comparison experiments are carried out.(4)Simulated roadway experiments and ground experiments were carried out,and the SLAM and navigation algorithms proposed in this thesis were tested and analyzed.The results show that the SLAM algorithm based on tightly coupled IMU proposed in this thesis achieves better results,the mapping loopback is more accurate and less ghosting,the positioning accuracy is improved by more than 50% compared with the optimization period,and the positioning error is within 12 cm.Compared with the improved scheme,the proposed converged path planning scheme improves the average computing efficiency and path cost by 57.62% and 7.81%.The thesis has 69 figures,22 tables,and 104 references.