一类基于输出非线性量测的变量带误差系统的辨识

IDENTIFICATION FOR A CLASS OF ERRORS-IN-VARIABLES SYSTEMS WITH NONLINEAR OUTPUT MEASUREMENTS

研究一类基于输出非线性量测的变量带误差系统的辨识.通过对输出量测的截断,在适当的系统假设下,应用扩张截断随机逼近算法给出了系统参数的递推估计,并证明了估计的强一致.辨识算法适用于多种常见的非线性量测.最后给出了一个仿真例子,仿真结果与理论一致.

This paper is concerned with the identification for a kind of errors-in-variables systems with nonlinear output measurements. With the measured outputs being truncated, recursive estimates for the system parameters are derived by the stochastic approximation algorithm under reasonable conditions. And the strong consistency for the estimates is established. The proposed algorithms are applicable for lots of common nonlinear observations. A simulation example shows the effectiveness of the theoretical analysis.