Research On Path Planning Algorithm For Robot In Unknown Environment

Path planning is an important aspect of the robotics research and it is also one ofthe most important task of robot navigation. Path planning is defined as: in theenvironment with obstacles, according to a certain criteria, the robot can find a pathfrom the starting position to the target position without collision. This topic takesmobile robot as the research object and analysis some algorithms of the mobile robotpath planning. This paper researches the artificial potential field method andrapidly-exploring random tree (RRT) algorithm deeply. According to the thelimitations and the defect in the application of two kinds of method, This paper putsforward the corresponding improvement methods, and the simulation proved thevalidity and reliability of the improved algorithm.The main parts of the paper include the following sections:Firstly, this paper introduces the research background of the robot path planningproblem and the purpose and meaning of this topic.The development situation isreviewed i at home and abroad. The path planning methods are contrasted. We analysistheir advantages and put forward the existing deficiency at the same time.Then, the traditional path planning methods are introduced in detail, and the pathplanning problems are analysised in the classical artificial potential field. Thetraditional artificial potential field method has the problem of Goals Nonreachablewith Obstacle Nearby (GNRON) and local minimum.This paper adopts improvedrepulsive potential function for GNRON of potential field. For the problem of localminimum, the paper improves the method of adding random force and adopting theequipotential line method. The improved method optimizes the path of robot avoidingobstacles. The improved method optimized the robot path without obstacle.In addition,the robot can also get out of the local minima with setting the temporary subtitlepunctuation.The simulation also can prove the efficiency of the new changing.Finally, the basic principle of RRT algorithm was introduced in detail, and thispaper analyzes the advantages and disadvantages of RRT algorithm and the existing improvement algorithms. This paper adopts the rolling windows mode aiming at therandom of RRT algorithm.The goal function of gravity of artificial potential fieldmethod is introduced into the RRT algorithm and this can keep the path optimal. Notonly can the mixture algorithm make up for the artificial potential field method in thedefect of local minima,but also it can make RRT algorithm get the path planningoptimal in the case of unknown environment.