Identification of a weighted combination of multivariable state space systems from input and output data

V. Verdult, M.H.G. Verhaegen

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    37 Downloads (Pure)

    Abstract

    Discusses a method for the determination of a weighted combination of local linear state-space systems from input and output data. The method is iterative and each iteration consists of two steps. The first step is to determine the weighting functions given the local models. This problem is solved by using an extended Kalman smoother. The second step is to identify the local models given the weights. For this step we optimize a cost function that represents a tradeoff between local and global learning. For this optimization we use a gradient search method in combination with an appropriate projection in the parameter space to deal with similarity transformations
    Original languageUndefined
    Title of host publicationProceedings of the 40th IEEE Conference on Decision and Control
    Place of PublicationOrlando, Florida, USA
    PublisherIEEE CONTROL SYSTEMS SOCIETY
    Pages-
    Number of pages6
    ISBN (Print)0-7803-7063-5
    DOIs
    Publication statusPublished - 4 Dec 2001
    Event40th IEEE Conference on Decision and Control, CDC 2001 - Hyatt Regency Grand Cypress, Orlando, United States
    Duration: 4 Dec 20017 Dec 2001

    Publication series

    Name
    PublisherIEEE Control Systems Society

    Conference

    Conference40th IEEE Conference on Decision and Control, CDC 2001
    Abbreviated titleCDC
    CountryUnited States
    CityOrlando
    Period4/12/017/12/01

    Keywords

    • METIS-205354
    • iterative methods
    • Smoothing methods
    • covariance matrices
    • Kalman filters
    • Jacobian matrices
    • State estimation
    • multivariable systems
    • State-spacemethods
    • Parameter estimation
    • nonlinear filters
    • Linear systems
    • IR-37644

    Cite this

    Verdult, V., & Verhaegen, M. H. G. (2001). Identification of a weighted combination of multivariable state space systems from input and output data. In Proceedings of the 40th IEEE Conference on Decision and Control (pp. -). Orlando, Florida, USA: IEEE CONTROL SYSTEMS SOCIETY. https://doi.org/10.1109/.2001.980959
    Verdult, V. ; Verhaegen, M.H.G. / Identification of a weighted combination of multivariable state space systems from input and output data. Proceedings of the 40th IEEE Conference on Decision and Control. Orlando, Florida, USA : IEEE CONTROL SYSTEMS SOCIETY, 2001. pp. -
    @inproceedings{5c4d68b4d415441fb1107a4927bbc094,
    title = "Identification of a weighted combination of multivariable state space systems from input and output data",
    abstract = "Discusses a method for the determination of a weighted combination of local linear state-space systems from input and output data. The method is iterative and each iteration consists of two steps. The first step is to determine the weighting functions given the local models. This problem is solved by using an extended Kalman smoother. The second step is to identify the local models given the weights. For this step we optimize a cost function that represents a tradeoff between local and global learning. For this optimization we use a gradient search method in combination with an appropriate projection in the parameter space to deal with similarity transformations",
    keywords = "METIS-205354, iterative methods, Smoothing methods, covariance matrices, Kalman filters, Jacobian matrices, State estimation, multivariable systems, State-spacemethods, Parameter estimation, nonlinear filters, Linear systems, IR-37644",
    author = "V. Verdult and M.H.G. Verhaegen",
    year = "2001",
    month = "12",
    day = "4",
    doi = "10.1109/.2001.980959",
    language = "Undefined",
    isbn = "0-7803-7063-5",
    publisher = "IEEE CONTROL SYSTEMS SOCIETY",
    pages = "--",
    booktitle = "Proceedings of the 40th IEEE Conference on Decision and Control",

    }

    Verdult, V & Verhaegen, MHG 2001, Identification of a weighted combination of multivariable state space systems from input and output data. in Proceedings of the 40th IEEE Conference on Decision and Control. IEEE CONTROL SYSTEMS SOCIETY, Orlando, Florida, USA, pp. -, 40th IEEE Conference on Decision and Control, CDC 2001, Orlando, United States, 4/12/01. https://doi.org/10.1109/.2001.980959

    Identification of a weighted combination of multivariable state space systems from input and output data. / Verdult, V.; Verhaegen, M.H.G.

    Proceedings of the 40th IEEE Conference on Decision and Control. Orlando, Florida, USA : IEEE CONTROL SYSTEMS SOCIETY, 2001. p. -.

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    TY - GEN

    T1 - Identification of a weighted combination of multivariable state space systems from input and output data

    AU - Verdult, V.

    AU - Verhaegen, M.H.G.

    PY - 2001/12/4

    Y1 - 2001/12/4

    N2 - Discusses a method for the determination of a weighted combination of local linear state-space systems from input and output data. The method is iterative and each iteration consists of two steps. The first step is to determine the weighting functions given the local models. This problem is solved by using an extended Kalman smoother. The second step is to identify the local models given the weights. For this step we optimize a cost function that represents a tradeoff between local and global learning. For this optimization we use a gradient search method in combination with an appropriate projection in the parameter space to deal with similarity transformations

    AB - Discusses a method for the determination of a weighted combination of local linear state-space systems from input and output data. The method is iterative and each iteration consists of two steps. The first step is to determine the weighting functions given the local models. This problem is solved by using an extended Kalman smoother. The second step is to identify the local models given the weights. For this step we optimize a cost function that represents a tradeoff between local and global learning. For this optimization we use a gradient search method in combination with an appropriate projection in the parameter space to deal with similarity transformations

    KW - METIS-205354

    KW - iterative methods

    KW - Smoothing methods

    KW - covariance matrices

    KW - Kalman filters

    KW - Jacobian matrices

    KW - State estimation

    KW - multivariable systems

    KW - State-spacemethods

    KW - Parameter estimation

    KW - nonlinear filters

    KW - Linear systems

    KW - IR-37644

    U2 - 10.1109/.2001.980959

    DO - 10.1109/.2001.980959

    M3 - Conference contribution

    SN - 0-7803-7063-5

    SP - -

    BT - Proceedings of the 40th IEEE Conference on Decision and Control

    PB - IEEE CONTROL SYSTEMS SOCIETY

    CY - Orlando, Florida, USA

    ER -

    Verdult V, Verhaegen MHG. Identification of a weighted combination of multivariable state space systems from input and output data. In Proceedings of the 40th IEEE Conference on Decision and Control. Orlando, Florida, USA: IEEE CONTROL SYSTEMS SOCIETY. 2001. p. - https://doi.org/10.1109/.2001.980959