Detection of abnormal movement of industrial robots using image sequence

In this paper, a monitoring system for industrial robots working in factories is proposed. This system automatically detects different movement from reference and immediately reports to the operator. It has the advantage that there is no need to receive any signal from the robot controller because it only observes with a simple camera such as CCD. For the detection method, an eigenspace method is used which is excellent in compression of image data and calculation of the correlation among images. Parametric eigenspace method (PEM) is also used to detect abnormalities such as the speed of movement. Some experiments using 2-axis robot show the effectiveness of the proposed method.