Research On Robot Motion Planning Based On Line-Segment-Basedmap

With the rapid development of computer science and sensor technology,the mobile robot is developing towards intelligence.The basic requirement of intelligent mobile robot is autonomous motion planning,while the foundation of motion planning is map building.Map building and motion planning in unknown environment will have important significance to the development of intelligent mobile robotics.With sensors,mobile robot getsdata information,builds the environment map,and plans a path for its movement,so as to search for a shortest path between initial position and target position.Finally,the robot tracks the shortest path to the target position.In this paper,the problem of robot motion planning based onLine-Segment-Based mapin the obstacle environment is studied.A line-Segment-Based map is constructed incrementally from readings of laser sensors.Then the Voronoi diagramof the Line-Segment-Based map is generated with Fortune's Sweep Line algorithm.The Voronoi diagram is searched with Dijkstra'salgorithm to determine the shortest accessible path from theinitial configuration to the goal.Finally,a path-tracking control law with Line-of-Sight approach is designed to track the reference path to the goal.All parts arecloselyrelatedtoeachother,and the main research contents are as follows:An environmental map model based on line segments is established.With a laser sensor,the environmental information is obtained,and three key algorithms,Generate-Line,Line-Matching,and Line-Fitting are designed.The detected points are extracted to generateline segment.Then,the line segment is matched with the line segments in the global map.If the match is successful,a new line segment fitted by the two line segments is imported into the global map;if the match is failed,it shows that the line segment has not been detected before,the line segment is directly added to the global map.Finally,the map is updated.Based on the Line-Segment-Based map,the Voronoi diagram is generated,and the shortest path can be generated in the Voronoi diagram.The method of generating Voronoi diagram by Voronoi diagram is studied,and a fast generating algorithm is designed;based on the Voronoi diagram and the Dijkstra'salgorithm,the search algorithm of the reference path is designed.Finally,the related algorithms are verified by programming.After getting the reference path,Line-of-Sight(LOS)algorithm is used to track the reference path,it simplifies the path tracking into angle tracking.The PD controller is designed to track LOS angle,and finally the robot is controlled to move to the goal.Based on the modeling method of environment map,generation algorithm of Voronoi diagram,path search algorithm and tracking method,the motion planning of the robot in the obstacle environment is achieved through C++ programming,and it is verified by MobileSim software.The experimental results show that the motion planning system is effective and feasible.