A Kalman-Filter-Based Method for Pose Estimation in Visual Servoing

• Published in: IEEE transactions on robotics Vol. 26; no. 5; pp. 939 - 947
• Main Authors: Janabi-Sharifi, Farrokh, Marey, Mohammed
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.10.2010; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptation; Algorithms; Applied sciences; Approximation; Artificial intelligence; Cameras; Computer Science; Computer science; control theory; systems; control; Control systems; Control theory. Systems; Estimating; Estimation; Exact sciences and technology; Kalman filter (KF); Kalman filters; Linearization; Modelling and identification; Noise; Pattern recognition. Digital image processing. Computational geometry; pose estimation; Real time; Robot control; Robot kinematics; robotic manipulator; Robotics; Robots; Sampling; Solid modeling; Statistics; Trajectory; Visual; visual servoing; Computer vision; Statistical analysis; Adaptive algorithm; Kalman filter; Kalman filter (KF); Iterative method; Visual servoing; control; Function approximation; Orientation; Real time; Extended Kalman filter; Robotics; Manipulator; Posture; robotic manipulator; Sampling rate; pose estimation; System identification; Linearization; Adaptation
• ISSN: 1552-3098; 1941-0468
• DOI: 10.1109/TRO.2010.2061290

The problem of estimating position and orientation (pose) of an object in real time constitutes an important issue for vision-based control of robots. Many vision-based pose-estimation schemes in robot control rely on an extended Kalman filter (EKF) that requires tuning of filter parameters. To obtain satisfactory results, EKF-based techniques rely on "known" noise statistics, initial object pose, and sufficiently high sampling rates for good approximation of measurement-function linearization. Deviations from such assumptions usually lead to degraded pose estimation during visual servoing. In this paper, a new algorithm, namely iterative adaptive EKF (IAEKF), is proposed by integrating mechanisms for noise adaptation and iterative-measurement linearization. The experimental results are provided to demonstrate the superiority of IAEKF in dealing with erroneous a priori statistics, poor pose initialization, variations in the sampling rate, and trajectory dynamics.