A Method on Parameter Estimation of Regression-Type Nonlinear Systems

Abstract

This paper deals with the parameter estimation of regression-type nonlinear system model based on maximum a posteriori (MAP) estimation. This is equivalent to the minimization problem of the object function derived from a posteriori probability density function. It is shown that the minimizer based on Newton-method becomes to iterated extended Kalman filter (IEKF). A method on the gain adjustment of the IEKF is given so as to improve the performance of it. Then the parameter setting for the gain adjustment is given as the solution of a continuous-time algebraic Riccati equation (CARE). As a result, the proposed estimator has some effect of simulated annealing.
It is shown in numerical example that the proposed method indicates acceptable performance compared with conventional methods.