6-DEGREE-OF-FREEDOM HAND EYE VISUAL TRACKING WITH UNCERTAIN PARAMETERS

Algorithms for full 3-D robotic visual tracking of moving targets whos e motion is 3-D and consists of translational and rotational component s are presented. The objective of the system is to track selected feat ures on moving objects and to place their projections on the image pla ne at desired positions by appropriate camera motion. The most importa nt characteristics of the proposed algorithms are the use of a single camera mounted on the end-effector of a robotic manipulator (eye-in-ha nd configuration), and the fact that these algorithms do not require a ccurate knowledge of the relative distance of the target object from t he camera frame. The detection of motion is based on a cross-correlati on technique known as Sum-of-Squares Differences (SSD) algorithm. The camera model used introduces a number of parameters that are estimated on-line, further reducing the algorithms' reliance on precise calibra tion of the system. An adaptive control algorithm compensates for mode ling errors, tracking errors, and unavoidable computations delays whic h result from time-consuming image processing. Experimental results ar e presented to verify the efficacy of the proposed algorithms and to h ighlight the limitations of the approach. These experiments were perfo rmed using a multirobotic system consisting of Puma 560 manipulators.