A run-time estimate method of measurement error variance for Kalman estimator

Abstract

In this paper, an estimate method of control system measurement error variance for Kalman estimator at run-time is proposed. In general, control system's measurement error variance is measured [7] by some measuring instruments or tuned [7] after implementation of Kalman estimator, but the precise measuring the control system measurement error variance is too hard. Moreover, even though we could tune the measurement error variance value, it could not deal with changes of system measurement error variance because of the phenomena of aging of system sensors or unpredictable external disturbances. So we introduce a method to estimate measurement error variance of the system for Kalman estimator at run-time to improve the performance of the estimator. Also a simple simulation and experimental result will be presented to show the effect of the proposed method. ？2007 IEEE.