Logo

Applications of Artificial Intelligence Control for Parallel Discrete-Manipulators

Pasila, Felix and Alimin, Roche (2016) Applications of Artificial Intelligence Control for Parallel Discrete-Manipulators. [UNSPECIFIED]

[img] PDF
Download (1424Kb)
    [img]
    Preview
    PDF (paper - Roche)
    Download (2767Kb) | Preview
      [img]
      Preview
      PDF (cek plagiasi - Roche)
      Download (2421Kb) | Preview

        Abstract

        Parallel Discrete-Manipulators are a special kind of force regulated manipulators which can undergo continuous motions despite being commanded through a large but finite number of states only. Real-time control of such systems requires very fast and efficient methods for solving their inverse static analysis. In this paper, artificial intelligence techniques (AI) are investigated for addressing the inverse static analysis of a planar parallel array featuring ten three-state force actuators and two applications using 3D Massively Parallel Robots (MPRs) with one and two layers. In particular, the research method used simulation software and hardware testing with the case of parallel manipulator with two level discrete pneumatic actuators. Simulations with typical desired displacement inputs are presented and a good performance of the results compared to AI is obtained. The comparison showed that the parallel manipulator has the Root Mean Squared Error (RMSE) has less than 10% and can be used for controlling the ternary states of discrete manipulators via AI.

        Item Type: UNSPECIFIED
        Uncontrolled Keywords: discrete-manipulators; artificial intelligence; inverse static analysis; three-state actuators
        Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
        Divisions: Faculty of Industrial Technology > Mechanical Engineering Department
        Depositing User: Admin
        Date Deposited: 31 Aug 2016 17:35
        Last Modified: 05 Jul 2022 11:44
        URI: https://repository.petra.ac.id/id/eprint/17498

        Actions (login required)

        View Item