The Development Of AGV Automatic Navigation Control System Based On ROS

Automated Guided Vehicle is a key equipment for Flexible Manufacturing System,Flexible Handling System and Automated Storage System.It can effectively improve the production efficiency and reduce the production cost as an important bridge connecting logistics and production.Under the driving of "Industry 4.0" and "Made in China 2025",the requirements of AGV's automation and intelligence level become higher,so there is great significance to study the non-rail-based intelligent AGV with automatic navigation function.In this paper,a four wheel independent steering vehicle(4WIS-AGV)is taken as the research object.Robot Operating System(ROS),which is characterized by distributed computing,code reusability and modular design,is used to develop AGV automatic navigation control application in the face of increasing complexity and scale of robot application software.Firstly,through the simplification of the actual model,the AGV model description file is programmed by using of the Unified Robot Description Format file,the model description file will be used to automatic navigation application development,virtual platform construction and test visualization.Based on the ROS navigation framework and the actual characteristics of AGV,the modules and system workflow of AGV automatic navigation function are analyzed,and the key modules such as robot localization,global and local path planning are developed.Through the analysis of 4WIS-AGV's relevant parameters in different motion modes,the kinematic model which the AGV localization,local path planning and odometer calculation rely on is established,and then a control flow chart for different motion modes is build.The virtual platform of AGV was established by using Gazebo simulator.Based on the virtual environment,grid-based map was built by using the technology of Simultaneous Localization And Mapping(SLAM),and then the automatic navigation and multi-objective planning Function was simulated,the simulation results show that AGV can realize the automatic navigation and obstacle avoidance in the process of moving to the target.In addition,the actual experimental results show that the precision of environmental map which is build by SLAM can meet usage requirements,meanwhile,the function of automatic navigation and obstacle avoidance can be also achieved efficiently.Speed control command which is sent to the slave computer is analyzed and optimized by utilizing the interface between Matlab/Simulink and ROS.At the same time,since the AGV is a complex and nonlinear system in this paper,it is difficult to establish an accurate mathematical model.A fuzzy motion controller which does not depend on the exact mathematical model of the controlled object is designed with the help of Simulink.