STRONG TRACKING FILTERING OF NONLINEAR TIME-VARYING STOCHASTIC-SYSTEMS WITH COLORED NOISE - APPLICATION TO PARAMETER-ESTIMATION AND EMPIRICAL ROBUSTNESS ANALYSIS

The strong tracking filter (STF) proposed by Zhou et al. in 1992, whic h was developed for nonlinear systems with white noise, is extended to a class of nonlinear time-varying stochastic systems with coloured no ise. A new concept of 'softening factor' is introduced to make the sta te estimator much smoother; its value can be preselected by computer s imulations via a heuristic searching scheme. The STF is then used to e stimate the parameters of a class of nonlinear time-varying stochastic systems in the presence of coloured noise. The robustness against mod el uncertainty of the STF is thoroughly studied via Monte Carlo simula tions. The results show that the STF has strong robustness against mod el-plant parameter mismatches in the statistics of the initial conditi ons, the statistics of the process noise and the measurement noise, th e system parameters, and the parameters in the measurement noise model . To a great extent the STF can give bias-free parameter estimations, where the parameters may be randomly time varying with unknown changin g law.