Condition monitoring concept for industrial Robots

Industrial robots are used in production technology for a wide variety of tasks. The most frequently used type worldwide is the so-called vertical articulated arm robot, often designed with 5 or 6 axes. Due to their relative movement, the axes are tribological systems, they are subject to wear and tear and must be maintained regularly. An important aspect of maintenance is the inspection, which aims to assess the current state of wear and tear. This paper presents a concept for condition monitoring by means of self-tests for industrial robots. The basis is formed by MEMS-based vibration sensors, which are mounted on the axis joints. The vibration signals acquired during the self-test are analyzed in an Edge Gateway and the condition is classified using methods from the field of machine learning. The result of the classification and the features used for it are then sent to a cloud platform where they can be further analyzed. With this approach, service calls can be planned in advance and unplanned downtimes avoided. The article concludes with a critical discussion of the advantages and disadvantages of the presented concept and gives an outlook on still open research questions.