Title :
Pose estimation of moving objects from video sequences based on the unscented transformation
Author :
I. Jovandic;Z. Durovic;B. Kovacevic
Author_Institution :
Systems Control Department, Faculty of Electrical Engineering, University of Belgrade, Belgrade 11020, King Alexander Boulevard 73, Serbia and Montenegro. ijovandic@etf.bg.ac.yu
fDate :
6/28/1905 12:00:00 AM
Abstract :
A method for moving objects tracking from video sequences is presented in this paper. The coordinate transform from image frame into the world frame is a nonlinear one. The technique of unscented transformation is used for measurement noise statistics approximation based on estimated statistics of measurement noise in the image frame. The proposed method is illustrated through a differential drive mobile robot tracking example
Keywords :
"Video sequences","Mobile robots","Robot kinematics","Statistics","Kalman filters","Filtering","Orbital robotics","Wheels","Noise measurement","State-space methods"
Conference_Titel :
Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on
Print_ISBN :
1-4244-0332-4
DOI :
10.1109/WCICA.2006.1713796