Approach to an automatic prediction of the state of industrial manipulators using machine learning methods

Abstract

Due to intensive development of information technologies and the onset of 4th industrial revolution the number of robotic industries is steadily growing. The volume of production and the use of robots is also increasing. At the same time, the support and the management of digital production is being rapidly developing. The robotic systems are incapable of completely excluding a person from the technological chain, since they need timely maintenance and personnel working out the emergency situations. One of the solutions to reduce the risk of unexpected breakdowns is a predictive approach to the maintenance. The implementation of this approach is carried out using data analysis tools. This study presents the results of applying machine learning methods to analyze data from industrial robots in order to predict potential failures