Research On Vision Guided AGV Control Technology

Navigation and control technology is the key of achieving the intelligence of AGV.Developing the high-performance automated guided vehicle has a great theoreticalsignificance and engineering application value. The object of study is front-wheel driveand guided AGV with vision navigation. Control system is designed through combiningthe theoretical research with the technology development to improve its performance.The contents of image processing are discussed, focusing on how to obtain accurateinformation of guide line. A practical picture processing system is designed. In the paper,the source image is cut to reduce the amount of data. The task of image filtering iscompleted by the combination of mathematical morphology which are corrosion andexpansion. The goal of picture segmentation is realized through the way of optimalthreshold method and Roberts operator to extract the information of guided line edge.Finally, based on the method of least squares, the goal of path recognition is achieved.The hardware architecture distributed, software architecture processes and hybridarchitecture of AGV control system are designed through the theoretical analysis.The positioning problem of AGV is studied and analyzed, which rely on thestructure characteristics, navigation mode and working environment of the researchobject. The appropriate path planning method of AGV is designed by comparing thecharacteristics of them. This method is based on deliberating planning and reaction wayof planning, which belongs to the path planning based on behavior.Finally, trajectory tracking problem of AGV is discussed in setting up kinematicsmodel, and put forward to a new terminal sliding mode. Combined with Lyapunovmethod, trajectory tracking control law is designed.The theoretical method proposed in this paper has been transformed into practicalcontroller by LabVIEW and used in AGV system successfully. The control technologycan identify the route real-time in the case of light interference. The modules arecoordinated, so AGV can work between the target location flexibly and steadily, traceroute accurately and fast.