Learning to bag with a simulation-free reinforcement learning framework for robots

被引:0
|
作者
Munguia-Galeano, Francisco [1 ]
Zhu, Jihong [2 ]
Hernandez, Juan David [3 ]
Ji, Ze [4 ]
机构
[1] Univ Liverpool, Cooper Grp, Liverpool, England
[2] Univ York, Sch Phys Engn & Technol, York, England
[3] Cardiff Univ, Sch Comp Sci & Informat, Cardiff, Wales
[4] Univ Cardiff, Sch Engn, Cardiff, Wales
基金
英国工程与自然科学研究理事会;
关键词
reinforcement learning; robot learning; robotics;
D O I
10.1049/csy2.12113
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Bagging is an essential skill that humans perform in their daily activities. However, deformable objects, such as bags, are complex for robots to manipulate. A learning-based framework that enables robots to learn bagging is presented. The novelty of this framework is its ability to learn and perform bagging without relying on simulations. The learning process is accomplished through a reinforcement learning (RL) algorithm introduced and designed to find the best grasping points of the bag based on a set of compact state representations. The framework utilises a set of primitive actions and represents the task in five states. In our experiments, the framework reached 60% and 80% success rates after around 3 h of training in the real world when starting the bagging task from folded and unfolded states, respectively. Finally, the authors test the trained RL model with eight more bags of different sizes to evaluate its generalisability.
引用
收藏
页数:15
相关论文
共 50 条
  • [31] Emotion in reinforcement learning agents and robots: a survey
    Thomas M. Moerland
    Joost Broekens
    Catholijn M. Jonker
    Machine Learning, 2018, 107 : 443 - 480
  • [32] Integrating Sporadic Imitation in Reinforcement Learning Robots
    Richert, Willi
    Scheller, Ulrich
    Koch, Markus
    Kleinjohann, Bernd
    Stern, Claudius
    ADPRL: 2009 IEEE SYMPOSIUM ON ADAPTIVE DYNAMIC PROGRAMMING AND REINFORCEMENT LEARNING, 2009, : 193 - 198
  • [33] Learning Advanced Locomotion for Quadrupedal Robots: A Distributed Multi-Agent Reinforcement Learning Framework with Riemannian Motion Policies
    Wang, Yuliu
    Sagawa, Ryusuke
    Yoshiyasu, Yusuke
    ROBOTICS, 2024, 13 (06)
  • [34] Learning to Move an Object by the Humanoid Robots by Using Deep Reinforcement Learning
    Aslan, Simge Nur
    Tasci, Burak
    Ucar, Aysegul
    Guzelis, Cuneyt
    INTELLIGENT ENVIRONMENTS 2021, 2021, 29 : 143 - 155
  • [35] Reinforcement Learning-based Learning from Demonstrations for Collaborative Robots
    Li, W. D.
    2021 IEEE 17TH INTERNATIONAL CONFERENCE ON AUTOMATION SCIENCE AND ENGINEERING (CASE), 2021, : 1642 - 1647
  • [36] Embedding Reinforcement Learning in Simulation
    AboElHassan, Ayman
    Yacout, Soumaya
    2021 26TH IEEE INTERNATIONAL CONFERENCE ON EMERGING TECHNOLOGIES AND FACTORY AUTOMATION (ETFA), 2021,
  • [37] Quantile Constrained Reinforcement Learning: A Reinforcement Learning Framework Constraining Outage Probability
    Jung, Whiyoung
    Cho, Myungsik
    Park, Jongeui
    Sung, Youngchul
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [38] A framework for learning in humanoid simulated robots
    Colombini, Esther Luna
    Simoes, Alexandre da Silva
    Germano Martins, Antonio Cesar
    Matsurra, Jackson Paul
    ROBOCUP 2007: ROBOT SOCCER WORLD CUP XI, 2008, 5001 : 345 - 352
  • [39] Towards A Learning Framework for Dancing Robots
    Tholley, Ibrahim S.
    Meng, Qinggang
    Chung, Paul. W. H.
    2009 IEEE INTERNATIONAL CONFERENCE ON CONTROL AND AUTOMATION, VOLS 1-3, 2009, : 1581 - 1586
  • [40] A Modular Simulation Platform for Training Robots via Deep Reinforcement Learning and Multibody Dynamics
    Benatti, Simone
    Tasora, Alessandro
    Fusai, Dario
    Mangoni, Dario
    PROCEEDINGS OF 2019 3RD INTERNATIONAL CONFERENCE ON AUTOMATION, CONTROL AND ROBOTS (ICACR 2019), 2018, : 7 - 11