Positioning and navigation method of underground drilling robot for rock-burst prevention based on IMU-LiDAR tight coupling

Abstract

The drilling robot for rock-burst prevention is the key equipment for pressure relief in rock-burst mines, and its accurate map construction and stable navigation under complex working conditions are the basis and premise for realizing intelligent drilling operations. Based on the analysis of the causes of point cloud distortion of LiDAR and the defects of classical SLAM algorithm, a point cloud distortion correction method based on the IMU continuous time trajectory is proposed, a data fusion model of LiDAR and IMU is established, and the positioning and mapping process of drilling robot based on the IMU-LiDAR tight coupling is designed. A closed ramp model is built based on the characteristics of coal mine pressure relief roadways, and the simulation analysis of the mapping effect is conducted. The results show that the proposed mapping algorithm outperforms existing commonly used methods in terms of positioning accuracy and trajectory error. On this basis, a dynamic path planning method based on the improved artificial potential field and rapidly-exploring random tree is proposed, and a path planning and navigation fusion scheme suitable for drilling robot is designed. Two simulation motion scenarios are then designed, and the results indicate that the proposed path planning method has a better comprehensive performance in terms of average path length, average running time, and average number of generated nodes in both global and dynamic path planning. In order to further verify the practicality of the positioning and navigation method, multiple comparative experiments are conducted in a simulated roadway, ground experimental base and underground pressure relief roadway, and the results indicate that after tightly coupling IMU data with LiDAR data, the positioning accuracy of the proposed method is significantly improved and has a superior positioning performance in feature degradation scenarios. In addition, the planning path has better performance in terms of computational efficiency and cost. The results prove the feasibility and superiority of the proposed positioning and navigation method in various scenarios.

Keywords

• drilling robot for rock-burst prevention
• simultaneous localization and mapping
• inertial-lidar fusion
• positioning and navigation
• path planning