An online intelligent robust adaptive LSQR estimation method for LTI state space model

Abstract Regarding the low accuracy and instability of common online methods for estimating dynamic models in time domain, presence uncertainty system dynamics, sensor noise environmental disturbances, this area is still open further research. In paper, a new estimation method proposed based on robust meta‐heuristic adaptive LSQR (ORALSQR) simultaneous multi input/output linear model state variables. This algorithm used to solve output matrix equations least squares error problem. The presented algorithm, its iterative nature, searches answer subspace by using logic. addition, solving steps search domain size each iteration are intelligently determined method. an identification maneuver, estimates variables estimated Kalman filter, then next addition stability proof presented. Numerical results show more robustness compared other mentioned paper which contain LS RLS methods.