Soft Object Dexterous Manipulation Using Deep Reinforcement Learning

DOI
  • Promma Sornsiri
    Department of Mechanical Information Science and Technology, Kyushu Institute of Technology
  • Chumkamon Sakmongkon
    Department of Mechanical Information Science and Technology, Kyushu Institute of Technology
  • Hayashi Eiji
    Department of Mechanical Information Science and Technology, Kyushu Institute of Technology
  • Mowshowitz Abbe
    Department of Computer Science, The City College of New York

説明

Manipulation of objects is one of the basic tasks that has been studied for a long time in the robotic field. Many experiments were setting the environment and using Deep Reinforcement Learning to train robot arm to grasp various objects. Still, most of those objects are solid objects, whereas nowadays, robot arms are used for grasping soft objects as well. In this study, we develop object manipulation tasks in pybullet simulation, focusing on soft objects by using Deep Reinforcement Learning (DRL) based on Soft Actor-Critic and Proximal Policy Optimization algorithm which aims to make the robot able to grasp soft objects in the exact position with proper force that does not damage them.

収録刊行物

キーワード

詳細情報 詳細情報について

  • CRID
    1390015333257428736
  • DOI
    10.5954/icarob.2023.os13-5
  • ISSN
    21887829
  • 本文言語コード
    en
  • データソース種別
    • JaLC
    • Crossref
    • OpenAIRE
  • 抄録ライセンスフラグ
    使用不可

問題の指摘

ページトップへ