The real-time vision measurement of multi-information of the bridge crane’s workspace and its application

Abstract

The measurement of the workspace information of the bridge crane is the premise of its intelligent control and safety monitoring. Some existing vision systems can only obtain very limited information, like swing angle of the payload. In this paper, a monocular vision measurement method is proposed to acquire multi-information of the payload off-centered angle, the payload rotation angle and the obstacle height. A hierarchical calibration method is designed to divide the large lifting height of the crane into multi-layers. At each layer, four markers are fixed on a plate symmetrically to obtain the center position of hook via the Blob analysis and ellipse fitting algorithm. After calibration, two fitting equations of the payload lifting height and the pixel coordinate of vertical lifting center are established corresponding to the pixel distance between two markers and the height of each layer. On the base of the two fitting equations, the measurement models of the payload off-centered angle, the payload rotation angle and the obstacle height are established combining with the geometric relations in the crane’s workspace, respectively. Finally, a crane prototype is established to validate the effectiveness of the vision measurement models. Moreover, the strategies of auto-centering control and automatic obstacle avoidance are designed utilizing the vision measurement system. The corresponding experiments were carried out and achieved a good performance. The research results of this paper have practical significance to intelligent control and safety monitoring of the crane in future.