This paper addresses the phenomenon of sloshing and the issues that arise during liquid handling at visual inspection stations. The pharmaceutical industry, recently put under pressure by the pandemic, has long adopted modular solutions consisting mainly of robotic islands. This work focuses on a visual inspection island for glass vials and ampules called VRU. This machine uses robotic arms to optimize the inspection process and enables automated control of a wide range of products using image recognition techniques and AI algorithms. However, the handling of containers in the presence of liquids requires special precautions to avoid the occurrence of bubbles inside the fluid that can prevent the cameras from correctly capturing any defects present. The banal solution involves a drastic reduction in the speeds and accelerations to which the liquids are subjected. However, using appropriate techniques makes it possible to achieve performance values similar to those obtainable when manipulating solid materials. The developed algorithms were tested using multibody simulations in the Mathworks Simscape environment and then validated using a six-axis Fanuc robot. In this study, however, the analysis conducted aimed to determine the correlations between trajectories, laws of motion, and sloshing in containers handled at high speed in industrial applications. In this study a multibody model was developed using a CFD analysis. The container consisted of a glass vial for pharmaceutical uses containing a liquid inside. The results obtained from the CFD analysis allowed us to calibrate the multibody model for the next phase of optimization of the laws of motion to be followed by the manipulator.
Read full abstract