Uncalibrated Image-Based Visual Servoing Based On Joint Space And Image Moment

This paper studies the uncalibrated image-based visual servoing problem and proposes a new solution based on joint-image Jacobian matrix and image moment. Firstly, we provide a mathematical definition of Joint-image Jacobian Matrix from the perspective of space map. Then, an online estimator based on Kalman filter is intended to estimate joint-image Jacobian matrix in real time. In order to enhance the stability of image feature extraction and improve the performance of the whole system, we analyse characteristics of image moment feature and image point feature, and choose moment-based feature to construct a feature vector for physical applications. Finally, we finish single-eye-in-hand image-based visual servoing experiments without calibration on the physical robot system platform. The experimental results illustrate the effectiveness and robustness of the proposed scheme.