Failures of sequential Bayesian filters and the successes of shadowing filters in tracking of nonlinear deterministic and stochastic systems

Physical Review E - Statistical, Nonlinear, and Soft Matter Physics

Volumn 79, Issue 6, 2009, Pages

  Judd, Kevin ^a   Stemler, Thomas ^a  

^a UNIVERSITY OF WESTERN AUSTRALIA   (Australia)

Author keywords

[No Author keywords available]

Indexed keywords

BAYESIAN FILTERS; PARTICLE FILTER; SEQUENTIAL FILTER;

AIR FILTERS; NONLINEAR FILTERING; NONLINEAR SYSTEMS;

BAYESIAN NETWORKS;

EID: 66049151122     PISSN: 15393755     EISSN: 15502376     Source Type: Journal    
DOI: 10.1103/PhysRevE.79.066206     Document Type: Article

References (24)

1. A. H. Jazwinski, Stochastic Processes and Filtering Theory (Academic, New York, 1970), Vol. 64.
2. K. Judd and L. A. Smith, Physica D 151, 125 (2001). 10.1016/S0167- 2789(01)00225-1
3. D. Ridout and K. Judd, Physica D 165, 26 (2002). 10.1016/S0167-2789(02) 00376-7
4. R. E. Kalman, J. Basic Eng. 82, 35 (1960).
5. K. Judd, Physica D 183, 273 (2003). 10.1016/S0167-2789(03)00180-5
6. M. S. Arulampalam, S. Maskell, N. Gordon, and T. Clapp, IEEE Trans. Signal Process. 50, 174 (2002). 10.1109/78.978374
7. P. Del Moral, Theor. Probab. Appl. 40, 690 (1995). 10.1137/1140078
8. A. Doucet, S. Godsill, and C. Andrieu, Stat. Comput. 10, 197 (2000). 10.1023/A:1008935410038
9. D. B. Rubin, in Bayesian Statistics 3, edited by, J. Bernardo, M. H. DeGroot, D. V. Lindley, and, A. F. M. Smith, (Oxford University Press, New York, 1988).
10. F. Le Grand and N. Oudjane, Ann. Appl. Probab. 14, 144 (2004). 10.1214/aoap/1075828050
11. K. Ikeda, Opt. Commun. 30, 257 (1979). 10.1016/0030-4018(79)90090-7
12. W. R. Gilks and C. Berzuini, J. R. Stat. Soc. Ser. B (Stat. Methodol.) 63, 127 (2001). 10.1111/1467-9868.00280
13. J. Bröcker, U. Parlitz, and M. Ogorzałek, Proc. IEEE 90, 898 (2002). 10.1109/JPROC.2002.1015013
14. M. Davies, Int. J. Bifurcation Chaos Appl. Sci. Eng. 3, 113 (1993). 10.1142/S0218127493000076
15. K. Judd, Physica D 237, 216 (2008). 10.1016/j.physd.2007.08.017
16. K. Judd, J. Nonlinear Sci. 18, 57 (2008). 10.1007/s00332-007-9010-x
17. K. Judd and L. A. Smith, Physica D 196, 224 (2004).
18. H. E. Rauch, F. Tung, and C. T. Striebel, AIAA J. 3, 1445 (1965). 10.2514/3.3166
19. Consider a density p (x1, x2) zero everywhere, except on the set { (x1, x2) R2: x2 = x12 }, and is unimodal on this line with a maximum at (1, 1). The marginal density on x1 is unimodal with a maximum at x1 =1, but the marginal density on x2 can have two peaks at x2 =1 and x2 =0. Depending on how the density lies on the line, the peak at x2 =0 can be larger than the peak at x2 =1; hence, a marginal density can be misleading. Consequently, it is generally better to maximize a joint density to avoid this kind of situation.
20. K. Judd, Phys. Rev. E 67, 026212 (2003). 10.1103/PhysRevE.67.026212
21. T. Stemler and K. Judd, Physica D 238, 1260 (2009).
22. T. Stemler and K. Judd (unpublished)..
23. K. Judd, Phys. Rev. E 75, 036210 (2007). 10.1103/PhysRevE.75.036210
24. K. Judd and T. Stemler, Philos. Trans. R. Soc. London (to be published).