The Research On Mobile Robot Path Planning In Irregular Obstacle Environment

The research on mobile robots has been paid great attention to by researchers.With the requirement of life quality continue to increase, the application of mobilerobots in the production and life has become wider and wider. Path planning as animportant part of robots technology directly restricts the scope of the robotapplication.This thesis mainly focuses on the study of the mobile robots path planning inirregular obstacle environment and its prevalent problems, the main content are asfollows:Firstly, analyses the characteristics of various path planning methods on the basisof reading a lot of international and domestic literature, and found that the existingmethods can be improved in the environment with irregular obstacles. Throughresearch, found that it is difficult to use the existing search algorithm to find theoptimal path in irregular obstacle environment, for the reason that it’s hard to use anappropriate method to establish irregularly shaped obstacles’ exact analytical model.Secondly, a method to create a topological map called three-dimensional-likemap was proposed, and the detailed way to create three-dimensional-like map wasdescribed. The three-dimensional-like map is to give a high value to all grid pointsaccording to certain rules based on the ordinary grid map. In this kind of map, it’s notneed to get further information to avoid obstacles, but only use the information ofobstacles’ height during the path planning.Thirdly, the concepts of the downward direction and the transit point areproposed based on a three-dimensional-like map, and the calculation methods of thedownward direction and the transit point were given. The downward direction pointsout the direction to avoid obstacles successfully with the least cost like a road sign,and as a result, the planning process will not fall into traps. Transit point is a vertex ofobstacles according to a given starting point and destination point in the planningprocess. It can avoid to create obstacles’ analytic model if the path formed by transitpoints. A path planning method was designed, simulation results show that thismethod can work very well in irregular obstacle environment, and the shortfall is thatthe obtained path is not the optimal path.Finally, in order to find the optimal path of the three-dimensional-like map, the concept of sub-path was proposed, and the sub-path contains the optimal path. Inorder to find the path from a number of sub-optimal paths quickly and efficiently, theA*algorithm was Improved. The key elements of improving are to construct costfunction which can assess the value of the path and state space formed by the sub-pathand to design reasonable search process. Several examples were given to illustrate theplanning approach.