Research On Path Planning Of Wheeled Robot Based On ROS

With the development of science and technology,intelligent robots have become important partners in people's daily lives.However,robot technology still has many shortcomings,especially in path planning.Related scholars also challenged it and proposed many new methods to optimize traditional path planning algorithm.The traditional global path planning A* algorithm and the local path planning dynamic window approach algorithm can accomplish simple navigation tasks,but there are still some problems.For example,in the process of traditional A* algorithm planning path,the number of redundant nodes brought by equal probability expansion of nodes,resulting in an excessively high Openlist node occupancy rate;the inflection point of the feasible path increases the length of the path,leading to the relative optimality of the planned path;the traditional dynamic window approach algorithm ignores the narrow space of the global feasible path in the process of driving the robot.In response to these problems,the latest research programs at home and abroad were investigated and new methods to solve these problems were proposed.On the basis of the traditional A* algorithm,it integrates the jump point search method and introduces the idea of pruning;based on the cost function of the traditional dynamic window approach algorithm,the subset points of the global planning path are taken as the local target points,and the cost function of the distance and angle of the local target point is introduced.The research found that the A* algorithm that integrates jump point search can reduce the node occupancy rate of Openlist,introduces the idea of pruning to cut the path length,and adds the dynamic window approach algorithm of the cost function of the local target point distance and angle to drive the robot through a narrow space.In summary,the path planning based on the ROS wheeled robot studied in this paper mainly focuses on the following three aspects:1.Aiming at the problem of high node occupancy rate in openlist,this paper combines the jump point search method to change the node expansion mode.In the process of node expansion,nodes with special properties are added to openlist to reduce the occupancy rate of openlist based on A* algorithm.2.Aiming at the relative optimality of the path,according to the axiom of the shortest line between two points,a pruning method is proposed.At the inflection point of the global path,a straight path without obliquely crossing obstacles is used to replace the broken path with inflection point,so as to achieve the purpose of pruning and planning the path length3.Aiming at the problem of ignoring the narrow space on the global path,the subset points of the global planning path nodes are regarded as the local target points.Based on the cost function of the dynamic window approach,the distance and angle cost functions between the local target points and the end of the trajectory are added to achieve the goal of passing the narrow space of the global planning path in the robot motion.Based on the above research content,this paper proposes a path planning algorithm which combines jump point search method,cuts the path length of inflection point and introduces the distance and angle cost function of local target point.Unmanned applications in regular or irregular map models such as wheeled robot hotel delivery,room disinfection and computer room security,etc.It has achieved the purpose of reducing the repetition rate of human labor,protecting human health and safety,and reducing the virus carrying in the human working environment.The improved path planning algorithm has better economy and higher efficiency than the traditional path planning algorithm,so the improved path planning method has certain engineering value for the intelligent application of wheeled robots.