Estimation of Continuous-Time Nonlinear Systems by Using the Unscented Kalman Filter

DOI DOI Web Site 参考文献6件 オープンアクセス
  • Min Zheng
    Graduate School of Advanced Technology and Science, The University of Tokushima
  • Kenji Ikeda
    Institute of Technology and Science, The University of Tokushima
  • Takao Shimomura
    Institute of Technology and Science, The University of Tokushima

説明

In this chapter, direct estimation of the continuous-time systems from the sampled I/O data by using the UKF like algorithm is paid attention, and the Rotary Pendulum is provided to estimate the parameters of the continuous-time nonlinear system for demonstrating the validity of the UKF. Through the simulation and the experiment results, we found that, for the numerical simulation, system parameters have been almost exactly estimated, and from the experimental I/O data, system parameter has been estimated within one percent RRSE by using the UKF like algorithm. All the simulations were set up under the condition that the initial value is known. The estimation of initial states is very important for obtaining the correct estimates of the system parameters. However, for the practical plants, the initial state may not be measured because there is a dead zone of the sensor. If the initial state is unknown, the covariance of the initial state has to be set large, and it leads to low precision of the parameter estimation. Therefore, we are to propose a continuous-time model estimation method by using the UKF like algorithm, in which the initial state as well as the paramters are estimated, as a future research.

収録刊行物

参考文献 (6)*注記

もっと見る

詳細情報 詳細情報について

問題の指摘

ページトップへ