RESEARCH ON ROBOTIC FEEDING SYSTEM BASED ON MACHINE VISUAL POSITIONING AND TRACKING

摘要

Feeding desiccant or deoxidant into food can is a common production process in food manufacture industry. To meet the demand of automatic feeding process, the industrial robot with the machine visual positioning and tracking system which guides the robot to accurately grasp the packet of desiccant or deoxidant on the conveyor through coordinate transformation in real-time is introduced. The machine visual tracking system consists of two parts: machine visual positioning and linear tracking. First, the image processing methods, including preprocessing, segmenting, edge detecting, edge fitting and corner and center point extracting, are adopted to determine the position and orientation of the desiccant or deoxidant packet on the conveyor. And then the linear tracking technique informs the robot of the distance the packet moves along the conveyor. The position of the food can on the conveyor is also obtained by the linear tracking technique, which allows the robot to feed the packet into the food can in motion accurately. This robotic feeding system has been applied into the albumen powder production which is verified to be accurate, high-speed and high reliability.