The Research On Warehouse Handing AGV Positioning Technology Based On Laser Slam

With the rapid development of China's E-commerce and Courier Services,China's delivery quantity showed a blowout growth.Therefore,in recent years,warehouse handing AGV which is the most intelligent in the logistics equipment is continuously applied to unmanned warehouses.To achieve automation and unmanned of the storage and sorting,and thus greatly improve the efficiency of logistics operations.Positioning is an important prerequisite that AGV works normally in the warehouse.At present,most of AGV adopts the positioning method of fixed path with low autonomy and low flexibility.Therefore,this paper studies a warehouse handing AGV autonomous positioning method in semi-structured warehouse environment——Warehouse Handing AGV Positioning Technology Based on Laser Slam.The main contents of this paper include:1)Analyzing the present situation of AGV positioning guidance technology,mobile robot autonomous positioning method and robot Slam technology at home and abroad.A positioning method based on laser Slam is proposed for warehouse handing AGV working in semi-structured warehouse environment.2)Analyze and study the hardware structure,software system and main sensor of Pioneer-LX mobile robot.3)Mathematical modeling of AGV positioning system.Establishing the robot coordinate system.According to the motion structure and mechanical structure of the Pioneer-LX mobile robot,the odometer movement model is established.Analyze several map description methods and create a global map.4)Based on the traditional ICP algorithm,an improved non-iterative improved ICP algorithm is proposed,and clustering algorithm is combined to eliminate the wrong result,which realizes the fast and accurate matching between the real-time laser data and the global map.And design the EKF localization algorithm for the positioning of AGV in the process of path travel.For the positioning of the AGV during the parking process,the shelf feature of the laser data is extracted and matched with the shelf feature of the global map to achieve the position correction and precise positioning before the parking operation.5)Design and build the experimental environment.The improved ICP algorithm and EKF localization algorithm are experimentally verified,and the experimental results show that the proposed algorithm and method can realize the autonomous positioning of AGV in the process of path travel and the precise positioning in the process of parking.