Robust Kalman filter-based least squares identification of a multivariable system

摘要: A novel direct identification using the residual model of Kalman filter (KF) is proposed for multiple-input and multiple-output Box–Jenkins system formed signal disturbance models relating input output without any a priori knowledge statistics measurement noise corrupting output. To avoid non-linear optimisation, auto-regressive moving average (MA) approximated by high-order MA model, so that unknown parameters KF enter linearly. key property established, namely transfer matrix fraction description (MFD) two-stage method developed here. In stage 1, identified robust, computationally efficient least-squares to capture completely both models. 2, derived balanced reduction technique. The MFD property. performance scheme successfully evaluated on simulated physical systems.