Auxiliary Variable-Based Identification Algorithms for Uncertain-Input Models

This study presents two auxiliary variable-based identification algorithms for uncertain-input models. The auxiliary variable-based least squares algorithm can obtain unbiased parameter estimates by introducing suitable auxiliary variable vectors. Furthermore, an auxiliary variable-based recursive least squares algorithm is proposed to reduce the computational efforts. To validate the framework and algorithms developed, it has conducted a series of bench tests with computational experiments. The simulated numerical results/plots are consistent with the analytically derived results in terms of the feasibility and effectiveness of the proposed procedure.