Maximum likelihood principle and moving horizon estimation based adaptive unscented Kalman filter

被引:119
作者
Gao, Bingbing [1 ]
Gao, Shesheng [1 ]
Hu, Gaoge [1 ]
Zhong, Yongmin [2 ]
Gu, Chengfan [2 ]
机构
[1] Northwestern Polytech Univ, Sch Automat, Xian 710072, Shaanxi, Peoples R China
[2] RMIT Univ, Sch Engn, Bundoora, Vic 3083, Australia
基金
中国国家自然科学基金;
关键词
Unscented Kalman filter; Noise statistic estimation; Maximum likelihood principle; Moving horizon estimation; Expectation maximization technique; NONLINEAR STATE ESTIMATION; COST; UKF;
D O I
10.1016/j.ast.2017.12.007
中图分类号
V [航空、航天];
学科分类号
08 ; 0825 ;
摘要
The classical unscented Kalman filter (UKF) requires prior knowledge on statistical characteristics of system noises for state estimation of a nonlinear dynamic system. If the statistical characteristics of system noises are unknown or inaccurate, the UKF solution will be deteriorated or even divergent. This paper presents a novel adaptive UKF by combining the maximum likelihood principle (MLP) and moving horizon estimation (MHE) to overcome this limitation. This method constructs an optimization based estimation of system noise statistics according to MLP. Subsequently, it further establishes a moving horizon strategy to improve the computational efficiency of the MLP based optimization estimation. Based on above, a new expectation maximization technique is developed to iteratively compute the MLP and MHE based noise statistic estimation by replacing complex smoothed estimates with filtering estimates for further improvement of the computational efficiency. The proposed method can achieve the online estimation of system noise statistic and enhance the robustness of the classical UKF. The efficacy of the proposed adaptive UKF is demonstrated through simulations and practical experiments in the INS/GPS integrated navigation. (C) 2017 Elsevier Masson SAS. All rights reserved.
引用
收藏
页码:184 / 196
页数:13
相关论文
共 36 条
[1]  
[Anonymous], 2010, Time series analysis and its applications: with R examples
[2]   Identification of process and measurement noise covariance for state and parameter estimation using extended Kalman filter [J].
Bavdekar, Vinay A. ;
Deshpande, Anjali P. ;
Patwardhan, Sachin C. .
JOURNAL OF PROCESS CONTROL, 2011, 21 (04) :585-601
[3]   Huber's M-Estimation-Based Process Uncertainty Robust Filter for Integrated INS/GPS [J].
Chang, Lubin ;
Li, Kailong ;
Hu, Baiqing .
IEEE SENSORS JOURNAL, 2015, 15 (06) :3367-3374
[4]  
Chitralekha S.B., 2009, IFAC Proceed, V42, P804
[5]   Robust positioning technique in low-cost DR/GPS for land navigation [J].
Cho, Seong Yun ;
Choi, Wan Sik .
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2006, 55 (04) :1132-1142
[6]  
Cramer H., 1999, Mathematical methods of statistics, V9
[7]   Unscented filtering for spacecraft attitude estimation [J].
Crassidis, JL ;
Markley, FL .
JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, 2003, 26 (04) :536-542
[8]   Sigma-point Kalman filtering for integrated GPS and inertial navigation [J].
Crassidis, John L. .
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS, 2006, 42 (02) :750-756
[9]   MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].
DEMPSTER, AP ;
LAIRD, NM ;
RUBIN, DB .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38
[10]  
Gao B. B., 2017, INT J CONTR IN PRESS