Automatic Chip Detection Using Differnet

Abstract

Chips entwined with workpieces and accumulated on the mechanical chuck in turning processes are cleaned up by operators in the factories, and this process is needed to be automated. In this study, an automatic chip detection system for the autonomous mobile robot (AMR) with a vision sensor is developed using DifferNet which can be trained with only a small amount of normal data. The results showed that the developed model was able to correctly determine the presence or absence of chips on the chuck.

Keywords: Process automation, Machine learning, Chip removal, Turning, Image processing