access icon free Kullback–Leibler divergence for interacting multiple model estimation with random matrices

© The Institution of Engineering and Technology
Received 20/11/2014, Accepted 06/07/2015, Published 01/02/2016

The problem of interacting multiple model (IMM) estimation for jump Markov linear systems with unknown measurement noise covariance is studied. The system state and the unknown covariance are jointly estimated, where the unknown covariance is modelled as a random matrix according to an inverse-Wishart distribution. For the IMM estimation with random matrices, one difficulty encountered is the combination of a set of weighted inverse-Wishart distributions. Instead of using the moment matching approach, this difficulty is overcome by minimising the weighted Kullback–Leibler divergence for inverse-Wishart distributions. It is shown that a closed-form solution can be derived for the optimisation problem and the resulting solution coincides with an inverse-Wishart distribution. Simulation results show that the proposed filter outperforms the previous work using the moment matching approach.

Inspec keywords: covariance matrices; state estimation; Markov processes; random processes; inverse problems; filtering theory; minimisation; method of moments; measurement errors; statistical distributions; linear systems

Other keywords: random matrix; jump Markov linear system; system state estimation; filtering theory; weighted Kullback-Leibler divergence minimisation; unknown measurement noise covariance; interacting multiple model estimation; weighted inverse-Wishart distribution; IMM; optimisation problem; moment matching approach

Subjects: Optimisation techniques; Markov processes; Signal processing theory; Optimisation techniques; Linear algebra (numerical analysis); Markov processes; Filtering methods in signal processing; Linear algebra (numerical analysis)

References

    1. 1)
    2. 2)
    3. 3)
    4. 4)
      • 20. Gupta, A.K., Nagar, D.K.: ‘Matrix variate distributions’ (Chapman and Hall, London, 1999).
    5. 5)
    6. 6)
    7. 7)
      • 24. Akaike, H.: ‘Information theory and an extension of the maximum likelihood principle’. Proc. of the Second Int. Symp. on Information Theory, 1973, pp. 267–281.
    8. 8)
    9. 9)
    10. 10)
      • 29. Lan, J., Li, X.R.: ‘Tracking of extended object or target group using random matrix, part I: new model and approach’. Proc. of the Int. Conf. on Information Fusion, Singapore, July 2012, pp. 2177–2184.
    11. 11)
    12. 12)
    13. 13)
    14. 14)
    15. 15)
    16. 16)
    17. 17)
    18. 18)
    19. 19)
    20. 20)
    21. 21)
      • 21. Bishop, C.M.: ‘Pattern recognition and machine learning’ (Springer, 2006).
    22. 22)
      • 30. Lan, J., Li, X.R.: ‘Tracking of extended object or target group using random matrix, part II: irregular object’. Proc. of the Int. Conf. on Information Fusion, Singapore, July 2012, pp. 2185–2192.
    23. 23)
    24. 24)
    25. 25)
    26. 26)
      • 22. Ali, S.M., Silvey, S.D.: ‘A general class of coefficients of divergence of one distribution from another’, J. R. Stat. Soc., 1966, 28, (1), pp. 131–142.
    27. 27)
      • 1. Costa, O.L.V., Fragoso, M.D., Marques, R.P.: ‘Discrete-time Markov jump linear systems’ (Springer, New York, 2005).
    28. 28)
      • 2. Bar-Shalom, Y., Li, X.R., Kirubarajan, T.: ‘Estimation with applications to tracking and navigation’ (Wiley-Interscience, 2001).
    29. 29)
      • 10. Li, X.R., Jilkov, V.P.: ‘Survey of maneuvering target tracking. Part V: multiple-model methods’, IEEE Trans. Aerosp. Electron. Syst., 2004, 41, (4), pp. 1255–1321.
    30. 30)
    31. 31)
    32. 32)
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-spr.2015.0149
Loading

Related content

content/journals/10.1049/iet-spr.2015.0149
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading