Workpiece detection on CNC milling machine tools is very important task of NC programming. The main aims are: ?? d etecting the workpiece and the position on machine tool table, ?? d etecting the workpiece shape, ?? d etecting the workpiece origin (co called ?Zero point?). In this paper, the system for optical determination of the workpiece origin on the CNC machine tools is described. The purpose of our research is to develop a low-cost (inexpensive) system for non-contact deter Determination of workpiece origin is very important procedure in order to speed-up the set-up process on machine tools and NC. When the camera is installed on the machine, it is first necessary to calibrate the system, so that the relationship between the image coordinate system and machine coordinate system can be determine (extrinsic parameters). Intrinsic camera parameters (focal length, image format,...) are also determined in the calibration. Calibration procedure is using developed feed-forward neural networks. With this method the calibration procedure is simplified and the mathematical derivation of camera model is avoided. Learned neural network represents the camera calibration model. After neural network learning is complete, we can begin using the system for determining the workpiece origin. The image of workpiece is captured with the help of low-cost camera. With appropriate algorithms for digital image processing the information of interest is separated. In our case, we want to find an area that represents the workpiece zero point (an end point on the workpiece). The input in the learned neural network is the workpiece zero point position in image coordinate system and output of the neural network is the position of the workpiece zero point in the machine coordinate system. This developed system was through a number of tests proved to be reliable and suitable for use in practice. The system is implemented on a three-axis CNC milling machine Lakos 150 G. Optical system can be, without major modifications, used in other CNC machine tools and also on robotic systems. In the paper, working of system is illustrated with a practical example, which confirms the effectiveness of the implemented system in actual use on machine.

Username
Password