Kronecker canonical forms for perspective systems with applications to parameter identification

In this paper, we introduce canonical forms for perspective dynamical systems and illustrate their application to parameter identification with the aid of a single CCD camera as sensor. We show that the parameters in the canonical form can be identified uniquely using an Extended Kalman Filter (EKF) provided that the a priori estimates of the parameters are close to the true value. The main contribution of this paper is twofold: Parameterization of perspective dynamical systems with a given set of Kronecker indices and application of EKF technique to identify the parameters.