Application of Self-Location System Using a Floor of Random Dot Pattern to an Automatic Guided Vehicle

Application of the AGV (automatic guided vehicle) has broadened in factories and warehouses. Although most AGVs use magnetic tape for the guide path, the guided method has some problems. Once a path has been set, you require re-installation of the magnetic guide tape on a floor in order to reset the path. Also it is difficult to set intricately branched. For this reason, the application of AGV in the field of logistics, such as day-to-day transport route changes, has been difficult. In recent years, in order to solve these problems, several navigation methods using various self-location systems have been developed. We have been developed a self-localization system which uses dots patterns on a floor. At first we construct the floor on which there are random dots patterns and make a database of the all dot positions on the floor. A camera captures an image of the floor and several dots are cropped in the image. The cropped dots are matched in the database of all the dot positions on the floor and then we can detect position and direction of the captured image on the floor. We have set the floor which has dots pattern for detecting position at a factory where several forklifts move around and tire marks of those forklifts are attached. And an AGV using the developed self-localization system has run autonomously in the factory. In this paper we report the interim results of the experiment.