Decoupled EKF for simultaneous target model and relative pose estimation using feature points

In this paper, the problem of combined target model and relative pose estimation for position-based visual servoing is addressed. The target object is assumed to be stationary with at least 3 distinguishable feature points. The midpoint triangulation method and a rough estimation method are developed for initial estimation of the target model and relative pose of target in current robot end-effector frame. A novel decoupled extended Kalman Filter (EKF)-based online estimation algorithm is proposed to improve the target model and relative pose estimation simultaneously under continuous robot dynamic motion. This new method is robust to large initial estimation errors and provides consistent and accurate target model estimation for optimal pose estimation as required in position-based visual servoing. Experimental results are given to demonstrate the performance of the proposed method