A New Robot Body Calibration Method of SCARA Based on Machine Vision

With the development of industrial automation, SCARA robot is widely used in industrial production because of its flexible action and fast speed. There is a certain error between the theoretical model and the actual model of SCARA robot due to the inconsistent in production and assembly process, which makes the robot positioning accuracy not guaranteed. At the same time, although the traditional calibration method like using laser tracker has high calibration accuracy, but the equipment is expensive and the calibration time is too long. In this paper, a calibration method based on machine vision for the SCARA robot is proposed, which uses the position error model to ensure the positioning accuracy of the robot's left-hand and right-hand systems. The calibration software is easy to use and the process is almost in automation. The experimental results show that the calibrating time is shorter and the positioning accuracy can be guaranteed.