Motion Control for an Automated Guided Vehicle in the Factory with a Stereo Camera

Abstract

This paper presents a study of lineless control for an automated guided vehicle (AGV) by using a stereo camera and markers. This control is called visual servoing, and studies of vehicle robots using visual servoing are proposed nowadays. The usual visual servoing controls assume that the target can be observed clearly or the target cannot be instantaneously observed. When the AGV works in a factory, robustness to changes in brightness is important. On the other hand, moving control without the line on the floor is also required for an industrial transportation robot. With this background, we propose an AGV transport system without the line on the floor. The proposed guidance method has two modes, i.e., marker and nonmarker modes. The visual servo control of an AGV in the marker mode uses image-based control, and the visual servo control in the nonmarker mode uses position-based control. This algorithm uses three markers. The AGV moves from the start to the first marker and from the third marker to the goal controls in the marker mode. From the first marker to the second marker, the AGV must pass between cars. The AGV between these markers is controlled by image and position-based control. The usefulness of the proposed control system is demonstrated by experimental results under brightness disturbances.