Randomized approximations of the image set of nonlinear mappings with applications to filtering

Fabrizio Dabbene, Didier Henrion, Constantino Manuel Lagoa, Pavel Shcherbakov

    Research output: Contribution to journalConference article

    4 Citations (Scopus)

    Abstract

    The aim of this paper is twofold: In the first part, we leverage recent results on scenario design to develop randomized algorithms for approximating the image set of a nonlinear mapping, that is, a (possibly noisy) mapping of a set via a nonlinear function. We introduce minimum-volume approximations which have the characteristic of guaranteeing a low probability of violation, i.e., we admit for a probability that some points in the image set are not contained in the approximating set, but this probability is kept below a pre-specified threshold ε. In the second part of the paper, this idea is then exploited to develop a new family of randomized prediction-corrector filters. These filters represent a natural extension and rapprochement of Gaussian and set-valued filters, and bear similarities with modern tools such as particle filters.

    Original languageEnglish (US)
    Pages (from-to)37-42
    Number of pages6
    JournalIFAC-PapersOnLine
    Volume28
    Issue number14
    DOIs
    StatePublished - Jul 1 2015
    Event8th IFAC Symposium on Robust Control Design, ROCOND 2015 - Bratislava, Slovakia
    Duration: Jul 8 2015Jul 11 2015

    All Science Journal Classification (ASJC) codes

    • Control and Systems Engineering

    Cite this

    Dabbene, Fabrizio ; Henrion, Didier ; Lagoa, Constantino Manuel ; Shcherbakov, Pavel. / Randomized approximations of the image set of nonlinear mappings with applications to filtering. In: IFAC-PapersOnLine. 2015 ; Vol. 28, No. 14. pp. 37-42.
    @article{d66f2fb12eba451db5071b54fdfa101f,
    title = "Randomized approximations of the image set of nonlinear mappings with applications to filtering",
    abstract = "The aim of this paper is twofold: In the first part, we leverage recent results on scenario design to develop randomized algorithms for approximating the image set of a nonlinear mapping, that is, a (possibly noisy) mapping of a set via a nonlinear function. We introduce minimum-volume approximations which have the characteristic of guaranteeing a low probability of violation, i.e., we admit for a probability that some points in the image set are not contained in the approximating set, but this probability is kept below a pre-specified threshold ε. In the second part of the paper, this idea is then exploited to develop a new family of randomized prediction-corrector filters. These filters represent a natural extension and rapprochement of Gaussian and set-valued filters, and bear similarities with modern tools such as particle filters.",
    author = "Fabrizio Dabbene and Didier Henrion and Lagoa, {Constantino Manuel} and Pavel Shcherbakov",
    year = "2015",
    month = "7",
    day = "1",
    doi = "10.1016/j.ifacol.2015.09.430",
    language = "English (US)",
    volume = "28",
    pages = "37--42",
    journal = "IFAC-PapersOnLine",
    issn = "2405-8963",
    publisher = "IFAC Secretariat",
    number = "14",

    }

    Randomized approximations of the image set of nonlinear mappings with applications to filtering. / Dabbene, Fabrizio; Henrion, Didier; Lagoa, Constantino Manuel; Shcherbakov, Pavel.

    In: IFAC-PapersOnLine, Vol. 28, No. 14, 01.07.2015, p. 37-42.

    Research output: Contribution to journalConference article

    TY - JOUR

    T1 - Randomized approximations of the image set of nonlinear mappings with applications to filtering

    AU - Dabbene, Fabrizio

    AU - Henrion, Didier

    AU - Lagoa, Constantino Manuel

    AU - Shcherbakov, Pavel

    PY - 2015/7/1

    Y1 - 2015/7/1

    N2 - The aim of this paper is twofold: In the first part, we leverage recent results on scenario design to develop randomized algorithms for approximating the image set of a nonlinear mapping, that is, a (possibly noisy) mapping of a set via a nonlinear function. We introduce minimum-volume approximations which have the characteristic of guaranteeing a low probability of violation, i.e., we admit for a probability that some points in the image set are not contained in the approximating set, but this probability is kept below a pre-specified threshold ε. In the second part of the paper, this idea is then exploited to develop a new family of randomized prediction-corrector filters. These filters represent a natural extension and rapprochement of Gaussian and set-valued filters, and bear similarities with modern tools such as particle filters.

    AB - The aim of this paper is twofold: In the first part, we leverage recent results on scenario design to develop randomized algorithms for approximating the image set of a nonlinear mapping, that is, a (possibly noisy) mapping of a set via a nonlinear function. We introduce minimum-volume approximations which have the characteristic of guaranteeing a low probability of violation, i.e., we admit for a probability that some points in the image set are not contained in the approximating set, but this probability is kept below a pre-specified threshold ε. In the second part of the paper, this idea is then exploited to develop a new family of randomized prediction-corrector filters. These filters represent a natural extension and rapprochement of Gaussian and set-valued filters, and bear similarities with modern tools such as particle filters.

    UR - http://www.scopus.com/inward/record.url?scp=84992529406&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84992529406&partnerID=8YFLogxK

    U2 - 10.1016/j.ifacol.2015.09.430

    DO - 10.1016/j.ifacol.2015.09.430

    M3 - Conference article

    AN - SCOPUS:84992529406

    VL - 28

    SP - 37

    EP - 42

    JO - IFAC-PapersOnLine

    JF - IFAC-PapersOnLine

    SN - 2405-8963

    IS - 14

    ER -