Implementation of a collaborative robot application for closures' quality control

Among the processes that SACMI IMOLA s.c.r.l has internally, quality control certainly plays a fundamental role. In particular, Between all the automatic machines that the company produces, the CCM48, a continuous compression moulding machine able, though 48 pistons, to create over 2000 plastic closures / min with max closure diameter 38 mm, is under the lens. The intent to make the machine competitive on the market leads to the necessity to create an excellent product. This makes quality control of the latter, an aspect of fundamental importance. In this regard, today the closure's control quality procedure is made by two operators that manually, cooperate together. The task is characterized by: LOW frequency, the quality control analysis is realized only once a day on a batch; HIGH repetability, the complete procedure need to be iterate for each cap. Moreover, the operations that are performed by the operators, might be automated through: selection of proper vision sensors for images acquisition, computer vision's algorithms for defects detection and robotic product handling. This is precisely the work that has been carried out in this industrial thesis. The goal is to increase the level of automation in the inspection of the product to highlight possible defects in the process and therefore in the machine. For this purpose, as will be highlighted later, it has been evaluated the possibility to install a collaborative robot to perform this task. The main reason is due to the necessity to do not enormously modify the environment, considering to let the operator works with the robot in a shared workspace. All the source code used to simulate the environment is available on the personal gitHub account \url{https://github.com/EdoardoG94/ThesisFolder}.