Artificial Intelligence in Predicting Abnormal States in a Robotic Production Stand

Abstract

The aim of the described study is an engineering solution to the problem of the implementation of artificial intelligence methods in predicting abnormal, extremely emergency states in robotic production stands. This task results from the need to improve the operational reliability of automated and robotic production lines, thus rationalizing the utility and cost values of these lines. The available hardware solutions as well as the existing and newly introduced new procedures and IT platforms are described. In the hardware part of the work, electric servo drives and gears of a multi-chain tripod robot were used, configured with the Festo Automation Suite software, programmed with the KEBA controller and the developed KeStudio application program.