Research On The Application Of AGV Positioning And Path Planning

In recent years,with the rapid promotion of the production and transportation level of enterprises,the unmanned logistics and the accelerated integration of various technologies,the application of unmanned logistics technology has become more and more normal for factories,warehousing and other industries.Under the multi-sensor fusion autonomous navigation mode,the implementation of operation requirements of industrial unmanned vehicles,such as the self-planning of picking and unloading cargo and obstacle routing,has become the main focus and direction of current research.It is the key to realize the intelligent robot to ensure its ability to realize autonomous positioning and path planning in an unknown environment.This article mainly aims at the indoor autonomous localization and path planning problem of AGV,with two-dimensional lidar RPLIDAR A1 and odometer as sensors to collect environmental information.The AGV test vehicle platform is built for simultaneous localization and mapping based on the ROS system,and based on the existing map of robot path planning of the work environment.Aiming at the problem of positioning and mapping of AGV,the AGV system was modeled and analyzed,the related coordinate system and transformation relationship of the AGV system were described,and the motion model and lidar observation model of the AGV were established.In PF algorithm on the basis of introducing the RBPF algorithm,particle degradation on RBPF algorithm,computation and memory consumption,serious problems,put forward the improvement proposal distribution and resampling method,in order to improve the positioning performance of the algorithm simulation confirm the effectiveness.Through the analysis of the current mainstream environment construction graph model,raster map is selected as the way of map construction in this paper.Aiming at the path planning problem of AGV,based on the established map and combined with the actual working scene requirements of AGV,the path planning problem of AGV is divided into global and local path planning problems.In the global path planning,the artificial potential field method is chosen as the global path planning algorithm,and the defects of the traditional potential field method are solved by improving the algorithm.In the local path planning,the dynamic window method is used as the local path planning algorithm.Finally,the simulation results show the effectiveness of the improved algorithm.Finally,an experimental analysis of SLAM and path planning in the working environment of the AGV was carried out on the established AGV experimental platform.The positioning and mapping tests of corridors and rooms were carried out on the AGV platform,and the path planning tests of the AGV were carried out on the map environment of the built rooms.The performance was tested according to the actual operation of the experimental AGV robot.Finally,the experimental results prove the effectiveness and feasibility of the system design.