基于中心差分的风险敏感滤波算法及其应用

李海君; 赵国荣

doi:10.15892/j.cnki.djzdxb.2013.05.025

2013 , Vol. 33 >Issue 5: 169 - 172

DOI: https://doi.org/10.15892/j.cnki.djzdxb.2013.05.025

摘要

针对航天器定姿系统存在较大初始误差下使系统不稳定,甚至滤波发散的问题,提出基于中心差分的风险敏感滤波算法。该方法结合风险敏感滤波和中心差分滤波各自的优点,将风险敏感估计算子引入中心差分滤波器中,消除初始误差大而导致的滤波收敛速度慢,甚至发散现象,提高滤波器的鲁棒性。将该算法与EKF、UKF及CDKF算法进行比较,仿真结果表明新算法在大初始化误差条件下有较好的滤波效果。

关键词： 非线性滤波; 风险敏感滤波; 中心差分滤波; 姿态确定

本文引用格式

李海君 , 赵国荣 . 基于中心差分的风险敏感滤波算法及其应用[J]. 弹箭与制导学报, 2013 , 33(5) : 169 -172 . DOI: 10.15892/j.cnki.djzdxb.2013.05.025

Abstract

The large initial error can result in instability and filtering divergence of spacecraft attitude determination system, so the risk sensitive filtering algorithms based on central differential was presented. With risk-sensitive estimator applied to central difference filter, the new algorithm integrates advantages of both risk-sensitive filter and central difference filter, so filtering convergence speed has been improved, filtering divergence has been restrained, and robustness has been strengthened. Compared with EKF, UKF and CDKF, the filtering result of new algorithm is better for larger initial error.

Key words： nonlinear filtering; risk-sensitive filtering; central difference filtering; attitude determination

[an error occurred while processing this directive]

参考文献

[1]	Lefferts EJ, Markley FL, Shuster M D. Kalman filteringfor spacecraft attitude estimation[J]. Journal of Guidance,Control, and Dynamics, 1982,5 (5): 417- 429.
[2]	Crassidis JL, Markley F L. Unscented filtering for space-craft attitude estimation[J]. Journal of Guidance, Control,and Dynamics, 2003, 26 (4): 536- 542.
[3]	Crassidis JL. Sigma-point Kalman filtering for integratedGPS and inertial navigation[C]//AIAA Guidance, Naviga-tion, and Control Conference, 2005: 1981-2004.
[4]	Oshman Y, Carmi A. Attitude estimation from vector obser-vations using genetic-algorithm-embedded quaternion parti-cle filter[J]. Journal of Guidance, Control, and Dynam-ics, 2006, 29(4): 879-891.
[5]	George J, Terejanu G, Singla P. Spacecraft attitude estima-tion using adaptive gaussian sum filter[J]. Journal of theAstronautical Sciences, 2009, 57 (1/2): 31-45.
[6]	Linares R, Kumar V, Singla P, et al. Information theoreticspace object data association methods using an adaptivegaussian sum filter[J]. Advances in the Astronautical Sci-ences, 2011, 140: 665680.
[7]	Lam QM, Crassidis JL. Precision attitude determinationusing a multiple model adaptive estimation scheme[C]//Proceedings of the IEEE Aerospace Conference, 2007.
[8]	Lam QM, Crassidis JL. Evaluation of a multiple model a-daptive estimation scheme for space vehicle's enhanced nav-igation solution[C]//AIAA Guidance, Navigation, andControl Conference, 2007: 4690-4710.
[9]	Marschke JM, Crassidis JL, Lam QM. Attitude estima-tion without rate gyros using generalized multiple model a-daptive estimation[C]//AIAA Guidance, Navigation, andControl Conference and Exhibit, 2009.
[10]	RK Boe, 1 MR James, IR Peterson. Robustness andrisksensitive filtering[J]. IEEE Trans. Automat. Contr., 2002,47(3): 451-461.
[11]	H Zhang, L Xie, YC Soh. Risk-sensitive filtering, pre-diction and smoothing for discrete-time singular systems[J]. Automatica, 2003,39(1): 57–66.
[12]	Thrun S, Langford J, Verma V. Risk sensitive particle fil-ters: Advances in Neural Information Processing Systems14 [M]. Cambridge: MIT Press, 2002: 372-381.
[13]	James MR, Elliott R J. Risksensitive and risk neutralcontrol for continuoustime hidden markov models[J]. Journal of Applied Mathematics and Optimization, 1996, 34(1): 37-50.
[14]	K Ito, K Xiong. Gaussian filters for nonlinear filteringproblems[J]. IEEE Trans. Automat. Cont., 2000, 45(5):910-927.

Options

文章导航