Title: Online estimation of Image Jacobian Matrix for uncalibrated dynamic hand-eye coordination

Authors: Jianbo Su, Yanjun Zhang, Zhiwei Luo

Addresses: Department of Automation, Shanghai Jiaotong University, Shanghai, 200240, China. ' Department of Automation, Shanghai Jiaotong University, Shanghai, 200240, China. ' Bio-Mimetic Control Research Center, RIKEN Anagahora, Shimoshidami, Moriyama-ku, Nagoya, 463-0003, Japan

Abstract: This paper focuses on three different strategies for online estimation of the Image Jacobian Matrix (IJM) in uncalibrated robotic visual servoing, with the help of different scenarios of system configurations and coordination tasks that are prevalent in current research. The least square estimation method and the constant IJM policy are proposed for monocular visual feedback, while a Kalman filter-based method is proposed for a stereovision system. The PI control law and the optimal control theory are respectively adopted for coordination controllers to suit different control purposes. Extensive simulations and experiments are provided to evaluate performance of the proposed methods.

Keywords: calibration-free; Image Jacobian Matrix; IJM; Kalman filter; least-square estimation; robotic visual servoing; online estimation; uncalibrated hand-eye coordination; dynamic hand-eye coordination; robot control; coordination control; robot vision.

DOI: 10.1504/IJSCC.2008.019582

International Journal of Systems, Control and Communications, 2008 Vol.1 No.1, pp.31 - 52

Published online: 17 Jul 2008 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article