Smart Factory: Automated Calibration Operations

Abstract

Calibration of test systems is a crucial part of radar production to ensure highquality products. The calibration process comprises many repetitive steps performed manually by operators. The process is time-consuming, which impedes the valueadding product testing in the test systems. This thesis investigates the potential for automating manual calibration processes. Furthermore, it aims to analyse the impact this would have on the availability of the test systems and the calibration quality. This was done through a process split into three distinct steps. First, a literature study focusing on collaborative robot technologies. A current state analysis was then performed to examine the current calibration process. Lastly, empirical testing was conducted by programming a collaborative robot to perform manual calibration tasks, mainly focusing on connecting and disconnecting connectors. The findings indicate that the majority of the tasks in the calibration can be performed by a collaborative robot equipped with a laser sensor and custom grippers for manipulating the cables. Some types of connectors are more challenging to handle using automation, suggesting that the process might still require an operator for certain steps, if not redesigned. The anticipated change in availability for a fully automated solution is minimal, as manual calibration will not be conducted frequently enough. An automated solution can improve calibration quality by performing tasks with better precision than operators. Nevertheless, the use of a cobot will result in greater equipment wear. The automation within this type of production environment has proven challenging, but the potential benefits of successful implementation should not be overlooked.