Font Size: a A A

Research On AGV Navigation Technology Based On Laser SLAM

Posted on:2021-04-19Degree:MasterType:Thesis
Country:ChinaCandidate:Z H SongFull Text:PDF
GTID:2428330611492540Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
With the development of social economy,AGV has been widely used in manufacturing,logistics,catering services,food industry and special environmental operations.It can effectively improve production efficiency,reduce labor costs,and is safer and more reliable.However,the complex and changeable working environment puts forward higher requirements for the automation and intelligence of AGV.In this paper,the laser slam and path planning algorithm are studied,combined with ROS system,which is implemented in AGV autonomous navigation.First of all,the motion model and observation model of AGV are established.By comparing three kinds of map representations commonly used in robot map building and navigation,the grid map is selected to represent the environment map,and the updating principle and process of the grid map are analyzed and deduced theoretically.Secondly,the SLAM algorithm based on filtering idea is analyzed theoretically,and the simulation comparison between particle filtering algorithm slam and extended Kalman filtering algorithm slam is made by MATLAB,which verifies the advantages of particle filtering algorithm in AGV map construction and autonomous navigation,and optimizes the rbpf-slam process by using lidar scanning data.Then,aiming at the problem of path planning based on known map,Matlab is used to simulate and analyze the algorithm of a'star and Dijkstra to verify that the algorithm of a'star is smaller and more efficient than Dijkstra in searching path range.A path planning scheme combining the algorithm of a'star and the method of dynamic window(DWA)is given,which uses the algorithm of A?star to carry out global path planning,and then uses the method of dynamic window(DWA)Carry out local path planning.Finally,the AGV navigation system software is designed based on ROS system,the gazebo simulation environment is constructed,and the improved rpbf-slam algorithm is used for map construction.On the prototype of AGV car designed and manufactured by ourselves,AGV motion control and navigation experiment using laser slam are carried out,and realized autonomous navigation function.
Keywords/Search Tags:Automated Guided Vehicle(AGV), Laser Lidar, Simultaneous Localization And Mapping(SLAM), Path Planning
PDF Full Text Request
Related items