Robustness and risk-sensitive filtering

被引：95
作者：
Boel, RK
James, MR
Petersen, IR
机构：
[1] State Univ Ghent, Elect Engn Dept, Syst Grp, Flemish Fdn Sci Res, Ghent, Belgium
[2] Australian Natl Univ, Fac Engn & Informat Technol, Dept Engn, Canberra, ACT, Australia
[3] Australian Def Force Acad, Dept Elect Engn, Canberra, ACT 2601, Australia
来源：
IEEE TRANSACTIONS ON AUTOMATIC CONTROL | 2002年 / 47卷 / 03期
基金：
澳大利亚研究理事会;
关键词：
estimation; filtering; H infinity; minimax; risk-sensitive; robustness;
D O I：
10.1109/9.989082
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
This paper gives a precise meaning to the robustness of risk-sensitive filters for problems in which one is uncertain as to the exact value of the probability model. It is shown that risk-sensitive estimators (including filters) enjoy an error bound which is the sum of two terms, the first of which coincides with an upper bound on the error one would obtain if one knew exactly the underlying probability model, while the second term is a measure of the distance between the true and design probability models. The paper includes a discussion of several approaches to estimation, including H-infinity filtering.
引用
收藏
页码：451 / 461
页数：11
相关论文
共 33 条
[21]  
KUMAR P. R., 2015, Stochastic Systems: Estimation, Identification, and Adaptive Control
[22]  
MCENEANEY WM, 1996, UNPUB ROBUST HINFINI
[23]  
MOORE JB, UNPUB J MATH SYST ES
[24]  
Mortensen R., 1968, J OPTIMIZ THEORY APP, V2, P386, DOI 10.1007/BF00925744
[25]   FILTERING AND SMOOTHING IN AN H-INFINITY SETTING [J].
NAGPAL, KM ;
KHARGONEKAR, PP .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1991, 36 (02) :152-166
[26]  
NGUANG SK, 1995, UNPUB HINFINITY FILT
[27]  
Petersen IR, 1996, INT J ROBUST NONLIN, V6, P267, DOI 10.1002/(SICI)1099-1239(199605)6:4<267::AID-RNC232>3.0.CO
[28]  
2-3
[29]  
PRA PD, 1995, UNPUB SOME CONNECTIO
[30]  
SPEYER JL, 1992, PROCEEDINGS OF THE 31ST IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-4, P2293, DOI 10.1109/CDC.1992.371382
1234