Research And Implementation Of Intelligent Path Planning Algorithm For Mobile Robot Based On Typical Path Library

Path planning is one of the major problems which need to be solved when guiding mobile robots to accomplish missions.Path planning is the method of choosing an optimal path from one or several paths between the starting position and the destination according to the environment and the characteristics of the robot.When dealing with large-scale path planning,a large number of reasonable paths need to be generated quickly,but traditional algorithm is not very effective in this case.After analyzing,we find that there may be similar or partial similarities among the different scenarios in which the mobile robot is located.In similar areas,a robot takes similar patterns to move and choose its path,which is possible for us to use existing scene information and path information to generate a new path.In this paper,we focus on the path planning problem in similar scene and propose an intelligent path planning algorithm for mobile robots based on typical path.Based on the existing scene and path information,the algorithm analyzes the attributes of the path to generate a number of reasonable paths quickly in a new scene.The algorithm consists of scene modeling,measurement of scene similarity,extraction of similar path characteristics,generation path,path evaluation and other modules.The main work of this paper includes the following aspects:(1)To the problem of scene modeling,we analyzed the characteristic properties of similar scenes.Clustering algorithm is used to realize scene partitioning.The results are the main basis for the description of the scene.(2)To the problem of scene similarity calculation,we treat the scene as a polygon.Based on the idea of polygon vertex matching algorithm,we calculate the similarity with different scenes,and find out the maximum region matching sequence from the matching scene with the dynamic programming method.(3)To the problem of typical path library,this paper proposes the "three-stage" path construction model,and extract similar paths from similar regions.The algorithm uses knowledge from similar paths to determine the key point coordinates,referring to coordinate mapping relationships and robot characteristics.The key coordinate points form the typical path library.(4)To the problem of optimal path selection,this paper proposes the path evaluation function model based on a variety of factors.Through a large number of simulation experiments,the weights of the parameters are determined,and the optimal path is selected from the typical path library.This paper has developed the scene layout and the path planning experiment platform.The feasibility of the algorithm has been verified by a lot of simulation experiments.The results show that the algorithm we proposed for similar large scale path planning problem can significantly improve efficiency,solving practical problems.