Nonlinear event-based state estimation using sequential Monte Carlo approach
| Autor: | Shaikshavali Chitraganti, Mohamed Abdelmonim Hassan Darwish |
|---|---|
| Přispěvatelé: | Control Systems |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
0209 industrial biotechnology
Computer science Covariance matrix Stochastic process Monte Carlo method Estimator 02 engineering and technology 01 natural sciences 010104 statistics & probability Nonlinear system 020901 industrial engineering & automation 0101 mathematics Particle filter Algorithm Event (probability theory) |
| Zdroj: | 2017 IEEE 56th Annual Conference on Decision and Control, CDC 2017, 2170-2175 STARTPAGE=2170;ENDPAGE=2175;TITLE=2017 IEEE 56th Annual Conference on Decision and Control, CDC 2017 CDC |
| Popis: | State estimation of nonlinear stochastic system in the setting of event-based (EB) measurements is quite challenging, because the measurements are not available at each sampling period, but are available only when a certain pre-specified event occurs. Recently, a nonlinear EB state estimator using Sequential Monte-Carlo approach is proposed in [21], where the authors obtained a EB state estimator for a given threshold. In this work, the results of [21] are extended as follows. Firstly, an empirical relation between the EB threshold and the average communication rate is obtained. Then, the performance of the estimator is evaluated by comparing the approximate error covariance matrix with the posterior Cramér-Rao bound. In addition, the computational complexity is addressed using the equivalent flop measure. Finally, the effectiveness of the proposed approach is demonstrated using two simulation examples. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|