Research Of Mobile Robot Path Planning Based On Improved A* Algorithm And Artificial Potential Field Algorithm

This paper focuses on the path planning of the robot.Path planning,is a method when a robot in a complex environment,to enable it can complete a work autonomously,which to search a collision free,optimal route from the current position to the target location.The problem of path planning is one of the most important research fields,and it has become an important research direction of artificial intelligence.Firstly,this paper introduces the robotic definition,its development,application and significance of this research.And from the domestic and international research status of the path planning,the current commonly used methods,those two aspects to do a detailed introduction.This paper compares these methods,which provide a theoretical basis for the choicing planning method.Then,based on a handling robot which design autonomously,this paper analyses the kinematics model of the robot,and introduces the key technology of the robot.Through the introduction,we can be a preliminary understanding how the robot to walk and rotate.This paper describes the method,which transformed from the global path planning coordinate system to the local path planning coordinate system for the same point in the environment.Then this paper introduces the robot how to accomplish the task of identifying objects and avoiding obstacles by the cooperation of the host computer and the slave computer,which to provide theoretical and hardware support for application of algorithm.Then,because of the defects of the traditional global path planning method and local path planning method,this paper proposes an improved A* algorithm based on energy conservation,which eliminates the influence of the heuristic function on the evaluation function.And adds the smoothing part It belongs to global path planning method.And proposes an improved artificial potential field method,The repulsive force field and the gravitational field function are improved,the problems of traditional algorithm is solved.It belongs to local path planning method.The two improved methods are simulated and analyzed,to verify the feasibility and necessity of the improved algorithm.Finally,according to the shortcomings when using the global path planning method or local path planning method lonely,this paper proposes a hybrid algorithm.The algorithm uses the improved A* algorithm for planning firstly,gets the initial path.Then each inflection point of the initial path is followed as a sub goal pointand.The improved artificial potential field method is applied to path planning,and the final path is obtained.At last the rationality and necessity of using this hybrid algorithm are proved through simulation analysis.This hybrid algorithm based on improved A* algorithm and improved artificial potential field method is presented in this paper,can be applied to many occasions by the simulation results.Compared with the other used commonly algorithms,the hybrid algorithm can provide a better path for the robot.