Position and Attitude Control of Eye-In-Hand System by Visual Servoing using Binocular Visual Space

We propose the 3D position and attitude control method of Eye-in-Hand System by visual servoing using Binocular Visual Space. The position of the target is estimated based on the linear approximation between translational motion space and binocular visual space. The attitude of the target is also estimated by the linear approximation between rotational motion space and posture binocular visual space. The proposed method is robust to calibration error of camera angles, because it does not use camera angles to calculate feedback command for translational and rotational velocity. Simulation results are presented to demonstrate the effectiveness of the proposed method.