Research On Path Planning Method Of Crawler Robot In Coal Mine

Coal is an important industry related to the lifeline of national economy,but the work of coal mining industry is hard and the potential risk is high.Therefore,it is of great significance to realize the intelligence and autonomy of underground mobile equipment in coal mine.And movement of ground mobile robot,the robot to realize the function of the independent walking,need to focus on for mining robot commonly used crawler travel mechanism,the characteristics of the coal mine tracked mobile robot path planning method,is put forward and test robust,fast search and have path smooth function of path planning algorithm,the concrete research content is as follows:The hardware and software of tracked mobile robot platform are researched and developed to meet the requirements of path planning.In the aspect of hardware,the hardware control frame is designed and the corresponding sensor is selected.Aiming at the serious problem of wireless signal absorption in coal mine roadway,the appropriate communication equipment was selected,and the maximum communication distance and the maximum transmission bandwidth of the system were verified by the communication test in the simulated roadway.In terms of software,URDF model is made according to Solid Works model,two kinds of tunnel scenarios are built in Gazebo simulation environment,and the robot handle control is realized based on turtle?bot node.In order to improve the path planning accuracy of coal mine robot,the research of 3D map construction was carried out.In order to meet the coal mine environment under the condition of application of robot in the dim light,weak texture,repetition and unstructured terrain built figure demand,research and comparison based on 3D laser radar graph algorithm,using KITTI data set of four kinds of scene contrast of graph based on NDT SLAM,graph based on ICP SLAM,and Le GO-LOAM three algorithms,select the optimal performance of Le GO-LOAM algorithm as the article built figure,and built figure tests have been carried out in the simulation of roadway.Aiming at the unstructured terrain environment and three-dimensional environment in the roadway of coal mine,the research of robot global path planning and local path planning and waypoint navigation are carried out.In the global path planning,the path planning method based on graph search was compared with that based on probability search.The path planning method based on searching the maximum probability path lookup table was adopted to design the trajectory of cubic spline curve based on the kinematics model of tracked robot.To solve the problem that the path is easy to fall into the local optimal value,a variable weight heuristic function based on Euclidean distance and search probability is designed,and jump points are used to improve the node exploration method.In the local path planning,the dynamic obstacle avoidance algorithm based on the trajectory occupation method was designed for the characteristics of laser point cloud degradation in the underground roadway.Combined with the soft optimization method,the trajectory of the robot was optimized to improve the curvature mutation of the trajectory,and the smoothness and smoothness of the trajectory of the robot was improved.Based on the autonomous navigation requirements of robot,the waypoint navigation framework of robot is constructed.In the simulation environment built,the proposed algorithm is tested based on 3D laser-point cloud map,and the path planning comparison test with Hybrid A* algorithm is carried out in the same scene.The waypoint navigation algorithm is tested in 2D and 3D environments.When there is no degradation of the mapping algorithm,the robot can meet the requirements of waypoint navigation.In order to evaluate the planning ability and comprehensive performance of the proposed path planning method,path planning tests were carried out on the crawler mobile robot platform developed in underground coal mine in the environment with weak texture,unstructured environment and simulated roadway.Path planning tests were also carried out on the drilling robot in underground coal mine.The experimental results show that the proposed path planning algorithm can work well in the weak texture and unstructured environment,and the robot can carry out stable path planning in the simulated tunnel.The search speed is millisecond level,the straight path has high straightness,and the turning path has high smoothness.It can make reasonable avoidance behavior to the dynamic obstacles in 2D and 3D environment,and realize the waypoint navigation function.The algorithm can also guide the drilling robot to walk autonomously.This paper has 124 figures,17 tables and 128 references.