A GPU parallel computing method for LPUSS

Chyon Hae Kim, Shigeki Sugano

    Research output: Contribution to journalArticle

    9 Citations (Scopus)

    Abstract

    We discuss the effective implementation of parallel processing for linear prediction-based uniform state sampling (LPUSS). In previous work, we proposed LPUSS as an optimization algorithm for mechanical motions that assures high optimality of the solutions and computational efficiency. In parallel computation, LPUSS requires balanced memory allocation and managed processing timing. In this paper, we propose an effective parallel computing method that assures high optimality and calculation efficiency in parallel processing using GPU processor. We conducted two experiments to validate the proposed method. In the first experiment, we compared single-thread processing for LPUSS and the proposed parallel processing. As a result of this experiment, calculation speed of LPUSS was about 4-20 times faster than that with single-thread CPU. In the second experiment, we applied the proposed method to the optimization of sixtuple inverted pendulum. As a result, the proposed method optimized the motion within 40 minutes. According to our survey, there is no other optimization method that is applicable to higher than quadruple inverted pendulum models with standard constraints.

    Original languageEnglish
    Pages (from-to)1199-1207
    Number of pages9
    JournalAdvanced Robotics
    Volume27
    Issue number15
    DOIs
    Publication statusPublished - 2013 Jul

    Fingerprint

    Parallel processing systems
    Sampling
    Processing
    Pendulums
    Experiments
    Storage allocation (computer)
    Computational efficiency
    Program processors
    Graphics processing unit

    Keywords

    • Dynamics
    • GPU
    • LPUSS
    • Motion planning
    • Parallel computing

    ASJC Scopus subject areas

    • Control and Systems Engineering
    • Human-Computer Interaction
    • Computer Science Applications
    • Hardware and Architecture
    • Software

    Cite this

    A GPU parallel computing method for LPUSS. / Kim, Chyon Hae; Sugano, Shigeki.

    In: Advanced Robotics, Vol. 27, No. 15, 07.2013, p. 1199-1207.

    Research output: Contribution to journalArticle

    Kim, Chyon Hae ; Sugano, Shigeki. / A GPU parallel computing method for LPUSS. In: Advanced Robotics. 2013 ; Vol. 27, No. 15. pp. 1199-1207.
    @article{47bb0bf5d2fc4be8aad15d652621849e,
    title = "A GPU parallel computing method for LPUSS",
    abstract = "We discuss the effective implementation of parallel processing for linear prediction-based uniform state sampling (LPUSS). In previous work, we proposed LPUSS as an optimization algorithm for mechanical motions that assures high optimality of the solutions and computational efficiency. In parallel computation, LPUSS requires balanced memory allocation and managed processing timing. In this paper, we propose an effective parallel computing method that assures high optimality and calculation efficiency in parallel processing using GPU processor. We conducted two experiments to validate the proposed method. In the first experiment, we compared single-thread processing for LPUSS and the proposed parallel processing. As a result of this experiment, calculation speed of LPUSS was about 4-20 times faster than that with single-thread CPU. In the second experiment, we applied the proposed method to the optimization of sixtuple inverted pendulum. As a result, the proposed method optimized the motion within 40 minutes. According to our survey, there is no other optimization method that is applicable to higher than quadruple inverted pendulum models with standard constraints.",
    keywords = "Dynamics, GPU, LPUSS, Motion planning, Parallel computing",
    author = "Kim, {Chyon Hae} and Shigeki Sugano",
    year = "2013",
    month = "7",
    doi = "10.1080/01691864.2013.813430",
    language = "English",
    volume = "27",
    pages = "1199--1207",
    journal = "Advanced Robotics",
    issn = "0169-1864",
    publisher = "Taylor and Francis Ltd.",
    number = "15",

    }

    TY - JOUR

    T1 - A GPU parallel computing method for LPUSS

    AU - Kim, Chyon Hae

    AU - Sugano, Shigeki

    PY - 2013/7

    Y1 - 2013/7

    N2 - We discuss the effective implementation of parallel processing for linear prediction-based uniform state sampling (LPUSS). In previous work, we proposed LPUSS as an optimization algorithm for mechanical motions that assures high optimality of the solutions and computational efficiency. In parallel computation, LPUSS requires balanced memory allocation and managed processing timing. In this paper, we propose an effective parallel computing method that assures high optimality and calculation efficiency in parallel processing using GPU processor. We conducted two experiments to validate the proposed method. In the first experiment, we compared single-thread processing for LPUSS and the proposed parallel processing. As a result of this experiment, calculation speed of LPUSS was about 4-20 times faster than that with single-thread CPU. In the second experiment, we applied the proposed method to the optimization of sixtuple inverted pendulum. As a result, the proposed method optimized the motion within 40 minutes. According to our survey, there is no other optimization method that is applicable to higher than quadruple inverted pendulum models with standard constraints.

    AB - We discuss the effective implementation of parallel processing for linear prediction-based uniform state sampling (LPUSS). In previous work, we proposed LPUSS as an optimization algorithm for mechanical motions that assures high optimality of the solutions and computational efficiency. In parallel computation, LPUSS requires balanced memory allocation and managed processing timing. In this paper, we propose an effective parallel computing method that assures high optimality and calculation efficiency in parallel processing using GPU processor. We conducted two experiments to validate the proposed method. In the first experiment, we compared single-thread processing for LPUSS and the proposed parallel processing. As a result of this experiment, calculation speed of LPUSS was about 4-20 times faster than that with single-thread CPU. In the second experiment, we applied the proposed method to the optimization of sixtuple inverted pendulum. As a result, the proposed method optimized the motion within 40 minutes. According to our survey, there is no other optimization method that is applicable to higher than quadruple inverted pendulum models with standard constraints.

    KW - Dynamics

    KW - GPU

    KW - LPUSS

    KW - Motion planning

    KW - Parallel computing

    UR - http://www.scopus.com/inward/record.url?scp=84886088562&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84886088562&partnerID=8YFLogxK

    U2 - 10.1080/01691864.2013.813430

    DO - 10.1080/01691864.2013.813430

    M3 - Article

    VL - 27

    SP - 1199

    EP - 1207

    JO - Advanced Robotics

    JF - Advanced Robotics

    SN - 0169-1864

    IS - 15

    ER -