Visual Servoing Control Based on EGM Interface of an ABB Robot

Abstract

This paper introduces a new visual servoing controller based on the (External Guided Motion, EGM) interface of ABB robots, with the emphasis on its adaptability to different feedback frequency without sacrificing cycle time and accuracy. Previous researches on visual servoing focused primarily on improving controllers range of stability and robustness to system errors. However, in industrial applications, besides the accuracy and stability of the system, the cost and the efficiency should also be taken into consideration. In practice, the cost of a camera is closely related to its frame rate. Hence, a visual servoing controller is expected to work with an image stream of low frame rate to save cost, while the speed and the accuracy of an operation are still maintained. With the feedback of robot's information from an External Guided Motion (EGM) interface, the performances of the visual servoing controller for the ABB robots under different camera frame rate can be maintained on an optimal level. The cycle times of picking operation and accuracies of tracking with different camera frame rate (from 16Hz to 200Hz) are compared in experiments to validate the effectiveness of the newly proposed controller.