Lennart Ljung

Publication List Details

Period

1974 - 2008

Number

74

Co-Authors

March 1987 LIDS-P- 16'60 OPTIMAL RECURSIVE HAXIMUM LIKELIHOOD ESTIMATION (2008)

Lennart Ljung, Sanjoy K. Mitter

Abstract: In this paper we derive stochastic differential equations for recursive maximum-likelihood estimates for the joint filtering-parameter estimation problem. Keywords: Maximum likelihood...

VINNOVA’s Competence Center ISIS Information Systems for Industrial Control and Supervision Annual Report 2004 (Public Version) (2008)

Lennart Ljung, Jonas Gillberg, Thomas Gustafsson, Marcus Klein, Erik Wernholt, Jonas Gillberg, ...

År 2004 har varit det första året av Fas IV. Under året har fördjupningen och utvidgningen av samarbetet fortsatt mellan de deltagande företagen och forskargrupperna vid LiTH (Linköpings...

comp.soft-sys.matlab Newsgroup (2008)

Lennart Ljung

ftp.mathworks.com Anonymous FTP server

A Basic Convergence Result for Particle Filtering (2008)

Hu, Xiao-Li, Schön, Thomas, Ljung, Lennart

The basic nonlinear filtering problem for dynamical systems is considered. Approximating the optimal filter estimate by particle filter methods has become perhaps the most common and useful method in...

Issues in closed-loop identification (2007)

Urban Forssell, Lennart Ljung

In this contribution we study the statistical properties of a number of closed-loop identification methods and parameterizations. A focus will be on asymptotic variance expressions for these methods....

Maximum Likelihood Identification of Wiener Models with a Linear Regression Initialization (2007)

Anna Hagenblad, Anna Hagenblad, Lennart Ljung, Lennart Ljung

Many parametric identification routines suffer from the problem with local minima. This is true also for the prediction-error approach to identifying Wiener models, i.e. linear models with a static...

A Comment On "leakage" In Adaptive Algorithms (2007)

Lennart Ljung, Jonas Sjöberg

. By "leakage" in adaptive control and adaptive signal processing algorithm is understood that a pull term towards a given parameter value is introduced. Leakage has been introduced both as...

A Parametric Method for Modeling of Time Varying Spectral Properties (2007)

Fredrik Gustafsson, Svante Gunnarsson, Lennart Ljung

The problem to track time-varying properties of a signal is studied. The somewhat contradictory notion of "time-varying spectrum" and how to estimate the "current" spectrum in an...

Variance Results for Closed-loop Identification Methods (2007)

Lennart Ljung, Lennart Ljung, Urban Forssell, Urban Forssell

In this contribution we study the statistical properties of a number of closed-loop identification methods and parameterizations. A focus will be on asymptotic variance expressions for these methods...

Shaping Frequency Dependent Time Resolution when Estimating Spectral Properties with Parametric Methods (2007)

Fredrik Gustafsson, Svante Gunnarsson, Lennart Ljung

The problem to track time-varying properties of a signal is studied. The somewhat contradictory notion of "timevarying spectrum" and how to estimate the "current" spectrum in an...

Report no.: LiTH-ISY-R-2139 For the IEEE Control Systems Magazine (2007)

Niclas Bergman, Niclas Bergman, Lennart Ljung, Lennart Ljung, Fredrik Gustafsson, Fredrik Gustafsson

Technical reports from the Automatic Control group in LinkSping are available by anonymous ftp at the address ftp.control.isy.liu.se. This report is contained in the pdf file 2139.pdf.

Maximum likelihood estimation of wiener models. Oct 2000. More info and ftp (2007)

Anna Hagenblad, Lennart Ljung

A Wiener model consists of a linear dynamic system followed by a static nonlinearity. The input and output are measured, but not the intermediate signal. We discuss the Maximum Likelihood estimate...

Interactive Analysis of Time-Varying Systems using Volume Graphics (2007)

Jimmy Johansson, David Lindgren, Matthew Cooper, Lennart Ljung, Jimmy Johansson, ...

Abstract — We show how 3-dimensional volume graphics can be used as a tool in system identification. Time-dependent dynamics often leave a significant residual with linear, timeinvariant models....

USING (2007)

Henrik Ohlsson, Jacob Roll, Torkel Glad, Lennart Ljung, Henrik Ohlsson, ...

A high-dimensional regression space usually causes problems in nonlinear system identification. However, if the regression data are contained in (or spread tightly around) some manifold, the...

Interactive Visualization as a Tool for Analysing Time-Varying and Non-Linear Systems (2005)

Jimmy Johansson, David Lindgren, Matthew Cooper, Lennart Ljung

This paper proposes advanced visualization and interaction techniques as a support for the analysis of system identification data. Non-linear or timedependent dynamics often leave a significant...

Terrain Navigation using Bayesian Statistics (1999)

Niclas Bergman, Lennart Ljung, Fredrik Gustafsson

this paper is therefore labeled the point-mass lter (PMF). 3 The Point-Mass Filter

Identification of Unstable Systems using Output Error and Box-Jenkins Model Structures (1999)

Urban Forssell, Urban Forssell, Lennart Ljung, Lennart Ljung

It is well known that for prediction error identification of unstable systems the output error and BoxJenkins model structures cannot be used. The reason for this is that the predictors in this case...

Optimal Recursive Maximum Likelihood Estimation, (1998)

Ljung, Lennart, Mitter, Sanjoy K., Moura, Jose M.

This paper derives stochastic differential equations for recursive maximum likelihood estimates for the joint filtering parameter estimation problem. Keywords: Maximum likelihood estimates;...

Closed-loop Identification Revisited (1998)

Urban Forssell, Urban Forssell, Lennart Ljung, Lennart Ljung

Identification of systems operating in closed loop has long been of prime interest in industrial applications. The problem offers many possibilities, and also some fallacies, and a wide variety of...

Identification for Control: Some Results on Optimal Experiment Design (1998)

Urban Forssell, Urban Forssell, Lennart Ljung, Lennart Ljung

The problem of designing the identification experiments to make them maximally informative with respect to the intended use of the model is studied. A focus will be on model based control and we show...

Identification for Control - What Is There To Learn? (1998)

Lennart Ljung, Lennart Ljung

This paper reviews some issues in system identification that are relevant for building models to be used for control design. We discuss how to concentrate the fit to important frequency ranges, and...

An Alternative Motivation for the Indirect Approach to Closed-loop Identification (1998)

Lennart Ljung, Lennart Ljung, Urban Forssell, Urban Forssell

Direct prediction error identification of systems operating in closed loop may lead to biased results due to the correlation between the input and the output noise. We study this error, what factors...

On Adaptive Smoothing of Empirical Transfer Function Estimates (1998)

Anders Stenman, Anders Stenman, Fredrik Gustafsson, Fredrik Gustafsson, Daniel Rivera, Daniel Rivera, ...

The determination of the right resolution parameter when estimating frequency functions for linear systems is a trade-off between bias and variance. Traditional approaches, like...

Point-mass filter and Cramer-Rao bound for terrain-aided navigation (1997)

Niclas Bergman, Lennart Ljung, Fredrik Gustafsson

The nonlinear estimation problem in navigation using terrain height variations is studied. The optimal Bayesian solution to the problem is derived. The implementation is grid based, calculating the...

Interpretation of subspace methods: Consistency analysis (1997)

Lennart Ljung, Lennart Ljung, Tomas Mckelvey, Tomas Mckelvey

Technical reports from the Automatic Control group in Linkoping are available by anonymous ftp at the address ftp.control.isy.liu.se. This report is contained in the compressed postscript file...

Point-mass filter and Cramer-Rao bound for Terrain-Aided Navigation (1997)

Niclas Bergman, Niclas Bergman, Lennart Ljung, Fredrik Gustafsson

The nonlinear estimation problem in navigation using terrain height variations is studied. The optimal Bayesian solution to the problem is derived. The implementation is grid based, calculating the...

Closed-loop Identification Revisited (1997)

Urban Forssell, Urban Forssell, Lennart Ljung, Lennart Ljung

In this contribution we study the statistical properties of a number of closed-loop identification methods and parameterizations. A focus will be on asymptotic variance expressions for these methods....

A Projection Method for Closed-loop Identification (1997)

Urban Forssell, Urban Forssell, Lennart Ljung, Lennart Ljung

A new method for closed-loop identification that allows fitting the model to the data with arbitrary frequency weighting is described and analyzed. Just as the direct method this new method is...

Nonlinear Black-Box Modeling in System Identification (1997)

Lennart Ljung

This paper describes the common framework for these approaches. It is pointed out that the nonlinear structures can be seen as a concatenation of a mapping from observed data to a regression vector...

A Projection Method for Closed-loop Identification (1997)

Urban Forssell, Urban Forssell, Lennart Ljung, Lennart Ljung

A new method for closed-loop identification that allows fitting the model to the data with arbitrary frequency weighting is described and analyzed. Just as the direct method this new method is...

Ensuring Certain Physical Properties in Black Box Models by Applying Fuzzy Techniques (1997)

Peter Lindskog, Lennart Ljung, P. Lindskog, L. Ljung

: We consider the situation where a nonlinear physical system is identified from input-output data. In case no specific physical structural knowledge about the system is available, parameterized grey...

Just in time models for dynamical systems (1996)

Anders Stenman, Fredrik Gustafsson, Lennart Ljung

The concept of just in time models is introduced for models that are not estimated until they are really needed. The idea is to store all observations of the process in a database, and then estimate...

Just in time models for dynamical systems (1996)

Anders Stenman, Fredrik Gustafsson, Lennart Ljung

The concept of just in time models is introduced for models that are not estimated until they are really needed. The idea is to store all observations of the process in a database, and then estimate...

Hybrid and discrete systems in automatic control – some new (Linköping) approaches (1996)

Lennart Ljung, Roger Germundsson, Inger Klein, Jonas Plantin, Jan-erik Stromberg

*:Currently at Stanford, y:Now with Ericsson Radio,z:Now with KTH, Stockholm The contributions from five recent Linkoping theses to the emerging, important and difficult field of hybrid and discrete...

Subspace Identification from Closed Loop Data (1996)

Lennart Ljung, Tomas Mckelvey

So called subspace methods for direct identification of linear models in state space form have drawn considerable interest recently. They have been found to work well in many cases but have one...

PAC-learning and Asymptotic System Identification Theory. (1996)

Lennart Ljung

In this paper we discuss PAC-learning of functions from a traditional System Identificaton perspective. The well established asymptotic theory for the identified models' properties is reviewed...

Just In Time Models For Dynamical Systems (1996)

Anders Stenman, Fredrik Gustafsson, Lennart Ljung

The concept of just in time models is introduced for models that are not estimated until they are really needed. The idea is to store all observations of the process in a database, and then estimate...

Aspects on Accelerated Convergence in Stochastic Approximation Schemes (1996)

Lennart Ljung

So called accelerated convergence is an ingenuous idea to improve the asymptotic accuracy in stochastic approximation (gradient based) algorithms. The estimates obtained from the basic algorithm are...

A Least Squares Interpretation of Sub-Space Methods for System Identification. (1996)

Lennart Ljung, Tomas McKelvey

So called subspace methods for direct identification of linear models in state space form have drawn considerable interest recently. The algorithms consist of series of quite complex projections, and...

Necessary and Sufficient Conditions for Stability of LMS (1996)

Lei Guo, Lennart Ljung, Guan-jun Wang

. In a recent work [7], some general results on exponential stability of random linear equations are established, which can be applied directly to the performance analysis of a wide class of adaptive...

Development of System Identification (1996)

Lennart Ljung

this paper are solid statistical ones. Otherwise, I think it must be said that the interaction between the statistics area and system identification field has been remarkably insignificant, in view...

Nonlinear black-box modeling in system identification: a unified overview (1995)

Jonas Sjoberg, Qinghua Zhang, Lennart Ljung, Albert Benveniste, Bernard Deylon, Pierre-yves Glorennec, ...

A nonlinear black box structure for a dynamical system is a model structure that is prepared to describe virtually any nonlinear dynamics. There has been considerable recent interest in this area...

Some aspects of nonlinear black box modeling in system identification (1995)

Lennart Ljung

The key problem in system identification is to find a suitable model structure, within which a good model is to be found. Fitting a model within a given structure (parameter estimation) is in most...

Classical Model Validation for Control Design Purposes (1995)

Lennart Ljung, Lei Guo

Model Validation is at the heart of the System Identification process. Recently, much renewed interest has been expressed in so called "identification for control". This means that the...

Necessary and Sufficient Conditions for Stability of LMS (1995)

Lei Guo, Lennart Ljung, G. J. Wang

. In a recent work [7], some general results on exponential stability of random linear equations are established, which can be appled directly to the performance analysis of a wide class of adaptive...

System Identification Through The Eyes Of Model Validation (1995)

Lennart Ljung, Håkan Hjalmarsson

Classical model validation procedures are placed at the focus of our attention. We discuss the principles by which we reach confidence in a model through such validation techniques, and also how the...

The Role of Model Validation for Assessing the Size of the Unmodeled Dynamics (1995)

Lennart Ljung, Lei Guo

The problem of assessing the quality of a given, or estimated model is a central issue in system identification. Various new techniques for estimating bias and variance contributions to the model...

A Tutorial On Multiple Model Least-Squares and Augmented UD Identification (1994)

Steve S. Niu, D. Grant Fisher, Lennart Ljung, Sirish L. Shah

The augmented UD identification (AUDI) is a family of new identification algorithms that are based on some well-known matrix decomposition and updating techniques. Compared with conventional...

Decomposition Methods for Least-Squares Parameter Estimation (1994)

Steve S. Niu, Lennart Ljung, Åke Björck

In this paper, least-squares method with matrix decomposition is revisited and a multiple model formulation is proposed. The proposed formulation takes advantage of the well-established decomposition...

Multiple Model Parameter Estimation (1994)

Steve S. Niu, Lennart Ljung

Handling many models simultaneously is a desired feature in least-squares estimation. This is typically handled by reducing the maximum order case to a triangular set of equations and then solving...

Efficient Construction of Transfer Functions from Frequency Response Data (1994)

Tomas McKelvey, Hüseyin Akçay, Lennart Ljung

In this paper, we present a novel non-iterative algorithm to identify linear time-invariant systems from frequency response data. The algorithm is related to the recent time-domain subspace...

Performance Analysis of General Tracking Algorithms (1994)

Lei Guo, Lennart Ljung

A general family of tracking algorithms for linear regression models is studied. It includes the familiar LMS (gradient approach), RLS (recursive least squares) and KF (Kalman filter) based...

Some Results on Identifying Linear Systems Using Frequency Domain Data (1993)

Lennart Ljung

The usefulness of frequency domain interpretations in linear systems is well known. In this contribution the connenctions between frequency domain and time domain expressions will be discussed. In...

A System Identification Perspecitve On Neural Nets (1992)

Lennart Ljung, Jonas Sjöberg

From a system identification perspective a neural net is "just another model structure". We review some of the basic system identification machinery to reveal connections with neural nets....

Black-box Identification of Transfer Functions: Asymptotic Results for Increasing Model Order and Data Records (1984)

LJUNG, LENNART, YUAN, ZHEN-DONG

The problem of estimating the transfer function of a linear stochastic system is considered. The transfer function is parametrized as a black box and no given order is chosen a priori. This means...