“A finite-difference method for linearization in nonlinear estimation algorithms”

Authors: Tor S. Schei,
Affiliation: SINTEF
Reference: 1998, Vol 19, No 3, pp. 141-152.

Keywords: Covariance matrices, estimation algorithms, extended Kalman filters, factorization methods, nonlinear filters, Jacobian matrices, linearization, parameter estimation, recursive estimation, state estimation

Abstract: Linearizations of nonlinear functions that are based on Jacobian matrices often cannot be applied in practical applications of nonlinear estimation techniques. An alternative linearization method is presented in this paper. The method assumes that covariance matrices are determined on a square root factored form. A factorization of the output covariance from a nonlinear vector function is directly determined by 'perturbing' the nonlinear function with the columns of the factored input covariance, without explicitly calculating the linearization and with no differentiations involved. The output covariance is more accurate than that obtained with the ordinary Jacobian linearization method. It also has an advantage that Jacobian matrices do not have to be derived symbolically.

PDF PDF (1581 Kb)        DOI: 10.4173/mic.1998.3.2

DOI forward links to this article:
[1] Changyuan Wang, Jing Zhang and Jing Mu (2012), doi:10.3390/s120708912
[2] Sina Razvarz, Raheleh Jafari and Alexander Gegov (2021), doi:10.1007/978-3-030-59246-2_8
[3] Manasi Das, Aritro Dey, Smita Sadhu and Tapan Kumar Ghoshal (2015), doi:10.1049/iet-smt.2014.0299
References:
[1] ANDERSON, B.D.O. MOORE, J.B. (1979). Optimal Filtering, Prentice-Hall, Englewood Cliffs, NJ.
[2] BIERMAN, G.J. (1977). Factorization Methods for Discrete Sequential Estimation, Academic Press, New York.
[3] CARLSON, N.A. (1973). Fast triangular formulation of the square root filter, A.I.A.A.J., 11, 1259-1265 doi:10.2514/3.6907
[4] CHENEY, W. KINCAID, D. (1985). Numerical Mathematics and Computing, Brooks/Cole, Monterey, CA.
[5] GELB, A. (1974). Applied Optimal Estimation, The MIT Press, Cambridge, MA.
[6] GOLUB, G.H. VAN LOAN, C.F. (1989). Matrix Computations, North Oxford Academic, London.
[7] JAZWINSKI, A.H. (1970). Stochastic Processes and Filtering Theory, Academic Press, New York.
[8] LJUNG, L. SÖDERSTRÖM, T. (1983). Theory and Practice of Recursive Identification, MIT Press, Cambridge, MA.
[9] PAPOULIS, A. (1984). Probability, Random Variables, and Stochastic Processes, McGraw-Hill, Singapore.
[10] THORNTON, C.L. BIERMAN, G.J. (1978). Filtering and error analysis via the UDUT covariance factorization, IEEE Trans. Automat. Control, AC-23, 901-907 doi:10.1109/TAC.1978.1101863


BibTeX:
@article{MIC-1998-3-2,
  title={{A finite-difference method for linearization in nonlinear estimation algorithms}},
  author={Schei, Tor S.},
  journal={Modeling, Identification and Control},
  volume={19},
  number={3},
  pages={141--152},
  year={1998},
  doi={10.4173/mic.1998.3.2},
  publisher={Norwegian Society of Automatic Control}
};