“Data and Program Structure for a Modular Extended Kalman Filter”

Authors: Ingar Solberg,
Affiliation: NTNU, Department of Engineering Cybernetics
Reference: 1988, Vol 9, No 4, pp. 179-189.

Keywords: Extended Kalman filter, nonlinear filtering, state space methods, computer programming

Abstract: The paper presents a data and program structure that makes it easier to implement a nonlinear process or measurement model when using the extended Kalman filter. This is achieved by a composite data type containing both the estimated value and covariance information. The basic operators ( +, -, *, /) and common functions are implemented for this data type. These enable a model for the extended Kalman filter to be implemented as easily as a discrete-time simulation model.

PDF PDF (937 Kb)        DOI: 10.4173/mic.1988.4.2

[1] BIERMAN, G.J. (1977). Factorization Methods for Discrete Sequential Estimation, Academic Press, New York.
[2] CARLSON, N.A. (1973). Fast triangular formulation of the square root filter, AIAA Journal, 11, 1259-1265 doi:10.2514/3.6907
[3] MACSYMA. (1988). Symbolics Inc, Four Cambridge Center.Cambridge, MA 02143.
[4] RALL, L.B. (1981). Automatic Differentiation: Techniques and Applications, Springer Verlag, Berlin.
[5] SOLBERG, I. (1988). A modular implementation of the extended Kalman filter with application to a crushing and screening circuit, Dr. Ing. thesis, Norwegian Institute of Technology, Division of Engineering Cybernetics, Trondheim.

  title={{Data and Program Structure for a Modular Extended Kalman Filter}},
  author={Solberg, Ingar},
  journal={Modeling, Identification and Control},
  publisher={Norwegian Society of Automatic Control}