Affiliation:
1. Northwestern Polytechnical University
Abstract
This paper reports the solution of the state estimation problem of nonlinear systems without knowing prior noise statistical characteristics. An adaptive UKF algorithm is proposed. This novel UKF algorithm is constructed by traditional UKF and EM algorithm, also it improve accuracy through Taylor series approximation. By applying the proposed algorithm into SINS/GPS integrated navigation system and comparing with the unscented Kalman filtering (UKF) algorithm, the adaptive UKF algorithm we proposed can effectively improve the filtering performance , also it outperforms UKF in terms of accuracy.
Publisher
Trans Tech Publications, Ltd.
Reference8 articles.
1. Zhao Lin, Wang Xiao-Xu, Sun Ming, Ding Ji-Cheng, YanChao. Adaptive UKF filtering algorithm based on maximum a posterior estimation and exponential weighting. Acta Au-tomatica Sinica, 2010, 36(7): 1007-1019.
2. J. Valappil, C. Georgakis, Systematic estimation of state noise statistics for extended Kalman filters, AIChE J. 46 (2) (2000) 292-308.
3. Den¾ux T. Maximum likelihood estimation from fuzzy data using the EM algorithm. Fuzzy Sets and Systems, 2011, 183(1): 72-91.
4. Xue Li, Gao shesheng, Zhao Yan. Unscented particle filtering with adaptive adjusted weight and its application in integrated navigation[J]. Journal of Chinese Inertial Technology. 2012, 20(4): 459-463.
5. G.C. Goodwin, J.C. Aguero, Approximate EM algorithms for parameter and state estimation in nonlinear stochastic models, 44th IEEE Conference on Decision and Control, 2005, p.368–373.
Cited by
4 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献