J. K. Tugnait, "Adaptive estimation and identification for discrete systems with Markov jump parameters," IEEE Trans. Automat. Contr., vol. AC-27, pp. 1054--1065, Oct. 1982.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....workshop proceedings) Relatively little attention, however, has been devoted to the study of the observability of both the continuous and discrete states of a hybrid system. To the best of our knowledge, the first attempt to characterize the observability of hybrid systems can be found in [8], although the condition given does not o#er much more insight than the definition itself. Observability has also been addressed recently in [6] which gives an unusual condition in terms of the existence of a discrete state trajectory. This condition pertains to systems where the discrete state ....

J. K. Tugnait. Adaptive estimation and identification for discrete systems with Markov jump parameters. IEEE Transactions on Automatic Control, 27(5):1054-- 1065, 1982.

....NSF STC CENS and ECS0200511, European Project RECSYS, MIUR National Project Identification and Adaptive Control of Industrial Systems and Italian Space Agency. 1. 1 Prior work To the best of our knowledge, the first attempt to characterize the observability of hybrid systems can be found in [15], where a definition of observability is proposed. 14] gives conditions for the observability of a particular class of linear time varying systems where the system matrix is a linear combination of a basis with respect to time varying coe#cients. 6] addresses the observability and ....

J. K. Tugnait. Adaptive estimation and identification for discrete systems with Markov jump parameters. IEEE Transactions on Automatic Control, 27(5):1054-- 1065, 1982.

....states. 1.2 Relation to prior work Filtering and identification of JLSs was an active area of research through the seventies and eighties: a review of the state of the art in 1982 can be found in [19] That paper discusses sub optimal algorithms for minimum mean square error state estimation. In [18], the same author uses a finite memory approximation to the maximum likelihood for simultaneously estimating the model parameters and the continuous and discrete states. The paper includes a condition for observability which, although tautological, is significant because it represents the first ....

J. K. Tugnait. Adaptive estimation and identification for discrete systems with Markov jump parameters. IEEE Transactions on Automatic Control, 27(5):1054--1065, 1982.

....Markov chain observed in white noise) to correlated noise. Under assumptions detailed later on, it is well known that exact computation of the conditional mean filtered or smoothed state estimates of and involves a prohibitive computational cost exponential in the (growing) number of observations [37]. This is unlike the standard HMM for which conditional mean state estimates can be computed with linear complexity in the number of observations via the HMM filter. Recently, efficient batch (off line) deterministic and stochastic iterative algorithms have been proposed to compute fixed interval ....

....to the prohibitive computational cost required to compute fixed lag and filtered state estimates of and ,itis necessary to consider in practice suboptimal estimation algorithms. A variety of algorithms has already been proposed in the literature to solve these estimation problems [4] 18] 36] [37]. Most of these algorithms are based on deterministic finite Gaussian mixture approximations like the popular Interacting multiple model (IMM) or the generalized pseudo Bayes (GPB) algorithms [4] These methods are computationally cheap, but they can fail in difficult situations. Another possible ....

[Article contains additional citation context not shown here]

J. K. Tugnait, "Adaptive estimation and identification for discrete systems with Markov jump parameters," IEEE Trans. Automat. Contr., vol. AC--25, pp. 1054--1065, 1982.

....the conditional probability density (or mass) function. Unfortunately, it is well known that exact computation of these estimates involves a prohibitive computational cost of order , where denotes the number of measurements and corresponds to all possible realizations of the finite Markov chain [25]. Thus, it is necessary to consider in practice suboptimal estimation algorithms. A variety of such suboptimal algorithms have been proposed; see, for example, 7] 24] and [25] In particular, 25] presents a truncated (approximate) maximum likelihood procedure for parameter estimation and a ....

...., where denotes the number of measurements and corresponds to all possible realizations of the finite Markov chain [25] Thus, it is necessary to consider in practice suboptimal estimation algorithms. A variety of such suboptimal algorithms have been proposed; see, for example, 7] 24] and [25]. In particular, 25] presents a truncated (approximate) maximum likelihood procedure for parameter estimation and a truncated approximation of the conditional mean state estimates. The estimates are computed using a bank of Kalman filters. This paper presents three stochastic iterative ....

[Article contains additional citation context not shown here]

J. K. Tugnait, "Adaptive estimation and identification for discrete systems with Markov jump parameters," IEEE Trans. Automat. Contr., vol. AC-25, pp. 1054--1065, 1982.

....sequence estimates of the finite state Markov chain and of the continuous state of the JMLS. Under assumptions detailed later on, it is well known that exact computation of these three estimates for JMLS involves a prohibitive computational cost exponential in the number, say , of observations [32]. Thus, it is necessary to consider in practice suboptimal estimation algorithms. A variety of such subopManuscript received January 25, 1999; revised February 20, 2001. The associate editor coordinating the review of this paper and approving it for publication was Dr. Joseph M. Francos. A. ....

....the Department of Mathematics, Statistics Group, University of Bristol, Bristol U.K. e mail: C.Andrieu bristol.ac.uk) Publisher Item Identifier S 1053 587X(01)03891 0. timal algorithms has already been proposed in the literature to solve these estimation problems [2] 11] 15] 23] 31] [32]. In this paper, we present original iterative stochastic and deterministic algorithms to perform MMSE and MMAP estimation of JMLS. The stochastic algorithms developed to estimate the MMSE and MMAP estimates are based, respectively, on homogeneous and nonhomogeneous Markov chain Monte Carlo (MCMC) ....

J. K. Tugnait, "Adaptive estimation and identification for discrete systems with Markov jump parameters," IEEE Trans. Automat. Contr., vol. AC-25, pp. 1054--1065, 1982.

....t observed in white noise) to correlated noise. Under assumptions detailed later on, it is well known that exact computation of the conditional 2 mean filtered or smoothed state estimates of x t and r t involves a prohibitive computational cost exponential in the (growing) number of observations [42]. This is unlike the standard HMM for which conditional mean state estimates can be computed with linear complexity in the number of observations via the HMM filter. Recently, e#cient batch (o# line) deterministic and stochastic iterative algorithms have been proposed to compute fixed interval ....

....computational cost required to compute fixed lag and filtered state estimates of x t and r t , it is necessary to consider in practice suboptimal estimation algorithms. A variety of algorithms have already been proposed in the literature to solve these estimation problems [5] 23] 41] [42]. Most of these algorithms are based on deterministic finite Gaussian mixture approximations like the popular Interacting Multiple Model (IMM) or the Generalised Pseudo Bayes (GPB) algorithms [5] These methods are computationally cheap but they can fail in di#cult situations. Another possible ....

[Article contains additional citation context not shown here]

J.K. Tugnait, "Adaptive estimation and identification for discrete systems with Markov jump parameters ", IEEE Trans. Automatic Control, vol. 25, pp. 1054-1065, 1982.

....probability density (or mass) function. Unfortunately, it is well known that exact computation of these estimates involves a prohibitive computational cost of order s T , where T denotes the number of measurements and s T corresponding to all possible realizations of the finite Markov chain [24]. Thus, it is necessary to consider in practice 1 sub optimal estimation algorithms. A variety of such suboptimal algorithms have been proposed, see for examples [7, 23, 24] In particular, 24] presents a truncated (approximate) maximum likelihood procedure for parameter estimation and a ....

....s T , where T denotes the number of measurements and s T corresponding to all possible realizations of the finite Markov chain [24] Thus, it is necessary to consider in practice 1 sub optimal estimation algorithms. A variety of such suboptimal algorithms have been proposed, see for examples [7, 23, 24]. In particular, 24] presents a truncated (approximate) maximum likelihood procedure for parameter estimation and a truncated approximation of the conditional mean state estimates. The estimates are computed using a bank of Kalman filters. This paper presents three stochastic iterative algorithms ....

[Article contains additional citation context not shown here]

J.K. Tugnait, "Adaptive Estimation and Identification for Discrete Systems with Markov Jump Parameters", IEEE Trans. Automatic Control, Vol. 25, pp. 1054-1065, 1982.

....Gaussian Observations together with martingale methods to derive a finite dimensional filter for s n . Instead of noisy measurements of the Markov chain X n , suppose that only noisy measurements of s n are available. In such a case, it is well known that the optimal state filter is infeasible [107]. Indeed the optimal state estimates would involve a computational cost that is exponential in the data length. Sub optimal finite dimensional approximations are given in [107] However, as we show in this chapter, given noisy observations y n of the Markov chain, the optimal state filter for s n ....

....only noisy measurements of s n are available. In such a case, it is well known that the optimal state filter is infeasible [107] Indeed the optimal state estimates would involve a computational cost that is exponential in the data length. Sub optimal finite dimensional approximations are given in [107]. However, as we show in this chapter, given noisy observations y n of the Markov chain, the optimal state filter for s n is finite dimensional. We also derive continuous time versions of the filters. Let us first describe the discrete and continuous time signal models. We then list the ....

[Article contains additional citation context not shown here]

J. Tugnait, Adaptive estimation and identification for discrete systems with Markov jump parameters, IEEE Trans. Auto. Control, 27 (1982), pp. 1054--1065.

....three estimates for JMLS involves a prohibitive computational cost, exponential in the number, say T , of observations. Thus in practice, it is necessary to consider suboptimal estimation algorithms. A variety of such suboptimal algorithms has already been proposed in the literature [9] 8] [12]. In this paper, we present iterative deterministic and stochastic algorithms to solve the three above mentioned problems. They have a computational cost of O (T ) per iteration. Convergence results are also obtained. The range of applicability of the proposed methods is wider than the previous ....

J.K. Tugnait, "Adaptive Estimation and Identification for Discrete Systems with Markov Jump Parameters", IEEE Trans. Automatic Control, vol. 25, pp. 1054-1065, 1982.

....observation model. The state estimation problem for discrete time Markov jump linear systems (without modal observations) has been studied in some detail both generally and in the context of tracking maneuvering targets (Ackerson and Fu, 1970; Jaffer and Gupta, 1971; Chang and Athans, 1978; Tugnait, 1982; Blom and Bar Shalom, 1988; Bar Shalom and Li, 1993; Costa, 1994) It is well known that the optimal filter is impractical due to exponential growth in computational and memory requirements (the filtered density is a Gaussian mixture and the number of terms grows exponentially with the data ....

Tugnait, J. (1982). Adaptive estimation and identification for discrete systems with Markov jump parameters.

No context found.

J. K. Tugnait, "Adaptive estimation and identification for discrete systems with Markov jump parameters," IEEE Trans. Automat. Contr., vol. AC-27, pp. 1054--1065, Oct. 1982.

No context found.

J. K. Tugnait. Adaptive Estimation and Identification for Discrete Systems with Markov Jump Parameters. IEEE Trans. Automatic Control, AC-27(5):1054--1065, October 1982.

No context found.

Tugnait, J.K.: Adaptive Estimation and Identification for Discrete Systems with Markov Jump Parameters. IEEE Trans. Automatic ControlAC-27 (1982) 1054--1065

No context found.

J. Tugnait, "Adaptive Estimation and Identification for Discrete Systems with Markov Jump Parameters," IEEE Trans. Automatic Control, AC-27(5):1054-- 1065, October 1982.

No context found.

Tugnait, J. K. (1982). Adaptive estimation and identification for discrete systems with Markov jump parameters, IEEE Trans. Automat. Control 27: 1054--1065. (correction: vol. 29, 1984, p. 286).

No context found.

J. K. Tugnait. Adaptive estimation and identification for discrete systems with Markov jump parameters. IEEE Transactions on Automatic Control, 27(5):1054--1065, Oct. 1982.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC