With the increase of labor cost and the development of technology,automatic handling of materials becomes the mainstream of industrial manufacture.Automatic guided vehicle(AGV)plays an important role in many working fields such as materials transportation and equipment docking.AGV is with high efficiency and precision and many other advantages so that its application scope and technology develops rapidly.Combining with related work of aluminum wheel casting workshop transformation project for an enterprise in Zhejiang Province,aiming at solving the shortage of tons of high temperature liquid aluminum material in the manual transport and filling process and meeting the requirements of improving working efficiency,a heavy-duty AGV system based on laser radar is designed to replace the manual to complete delivery with the filling of liquid aluminum material.The main content consists of following parts:1.We study a lot of AGV automatic guided technology and analyze the characteristics of several kinds of guidance technology in detail considering the application condition and the technical indicators.Then we propose overall technical scheme of the AGV system based on laser radar.2.We design a positioning navigation algorithm for AGV system.A special artificial landmark and a special recognition algorithm are designed.Then we use extend Kalman filter to reduce the error caused by the fluctuation of laser radar data.Using part of landmark information and RFID tag information as environmental feature information,a method of feature matching is used to calculate the pose deviation of AGV from the scheduled route.According to the motion model of AGV,we analyze the different positioning accuracy requirement of AGV in different operation areas and design the different motion control strategy.3.We develop a PC software based on MFC framework for this AGV.This software is developed in a modular programming way to achieve the main functions of AGV.The interaction interface of this software is programmed according to the operating process.4.A large number of experimental researches,including landmark recognition experiment,landmark pose estimation experiment,correction experiment of radar installation error and running experiment under simulated condition is conduct.The results show that the AGV system can realize autonomous navigation and the positioning is fast and accurate.We prove that the system running results meet the design requirements. |