Image mosaicing based condition monitoring approach for multi robots at production lines in industrial autonomy systems

Abstract

In today industry, manufacturing become big and serial as it never been before thanks to the autonomy robots. Hitches on such autonomy systems used in industrial production may cause production delaying. In this study, it is aimed to obtain alive bird's eye view map of full system in order to monitor manufacturing robots at production facilities that are big and impossible to be monitored with only one camera. Finding the similar scenes of input images, estimation of homography, warping and blending operations are applied respectively in order to mosaic the images by twos. Thus the robots in the facility can be observed in one screen. With observation of the obtained images, faults on cyber-physical systems that may cause damage in machines which are not cheap can be handled beforetime.