Error Compensation Method during Welding Operations using the Enhanced Wagner-Hagras Interval Type-3 NSFLS-1

Abstract

Automated welding with robots is one of the core processes of modern manufacturing. Industrial robots are capable of continuously performing welding tasks under different working conditions in well-defined and structured environments. However, reprogramming in the plant is still necessary to deal with uncertainties due to a positioning error of the parts to be welded on the production line. This error can be due to several reasons, including dimensional quality of the parts, mechanical play, aging of the mechanisms, etc. In this sense, an automatic decision-making system is required to compensate for the measured error and avoid stopping the production line. In this work, we propose a controller based on the Enhanced Wagner-Hagras Interval Type-3 NSFLS-1 (EWH Type-3 NSFLS-1), a vision system and a laser beam incident on the plates to be welded to determine the misalignment that is corrected by a fuzzy controller. Type-3 fuzzy logic systems (T3 FLS) make it possible to model the effects of uncertainties and to minimize them by optimizing the parameters during the learning process of the fuzzy system. They can approximate any real continuous function on a compact set to arbitrary accuracy. In this article, several fuzzy logic algorithms were evaluated, resulting in the proposed EWH IT3 TSK NSFLS-1 controller being the best option since it presented the lowest error of 1.79e-09m to successfully reposition the robot.