Title :
Visual motion estimation of 3D objects: an adaptive extended Kalman filter approach
Author :
Lippiello, Kncenzo ; Siciliano, Bruno ; Villani, Luigi
Author_Institution :
Dipartimento di Informatica e Sistemistica, Universita degli Studi di Napoli Federico II, Italy
fDate :
28 Sept.-2 Oct. 2004
Abstract :
An algorithm for the visual estimation of the pose of a moving object is presented in this paper. The algorithm exploits the prediction capability of the extended Kalman Filter to realize in real time a dynamic optimal selection of the object image features used for pose estimation. The robustness of the system with respect to the measurement noise and modelling errors is enhanced by using an adaptive scheme. Experimental case studies are presented to prove the effectiveness of the proposed approach.
Keywords :
adaptive Kalman filters; motion estimation; noise measurement; 3D object visual motion estimation; adaptive extended Kalman filter approach; noise measurement; pose estimation; Cameras; Covariance matrix; Filters; Motion estimation; Noise robustness; Optical sensors; Prediction algorithms; Robot kinematics; Robot vision systems; Statistics;
Conference_Titel :
Intelligent Robots and Systems, 2004. (IROS 2004). Proceedings. 2004 IEEE/RSJ International Conference on
Print_ISBN :
0-7803-8463-6
DOI :
10.1109/IROS.2004.1389476
