CONVERGENCE OF THE EKF IN MARK-BASED VISION FOR 3D VEHICLE TRACKING

摘要

In this paper, the extended Kalman filtering (EKF) technique is considered for 3-dimensional tracking of vehicle movement using a fixed camera that provides vehicle images showing several marks, easily detectable, fixed to the vehicle body. The algorithm will be implemented in a minihelicopter hover-stabilization application. For this problem, we present results on the convergence and the domain of attraction of the non-linear observation scheme as a function of the tuning filter parameters: the initial value of the covariance and the robustifying parameter α.