Enhancing learning capabilities of movement primitives under distributed probabilistic framework for flexible assembly tasks

被引:2
|
作者
Wang, Likun [1 ]
Jia, Shuya [2 ]
Wang, Guoyan [3 ]
Turner, Alison [1 ]
Ratchev, Svetan [1 ]
机构
[1] Univ Nottingham, Ctr Aerosp Mfg, Nottingham, England
[2] Bauman Moscow State Tech Univ, Sch Comp Sci & Syst, Moscow, Russia
[3] Safran Landing Syst, Mfg Dept, Gloucester, England
来源
NEURAL COMPUTING & APPLICATIONS | 2023年 / 35卷 / 32期
关键词
Learning from demonstration; Task-parametrised; Probabilistic distributed framework; Bayesian committee machine; Assembly;
D O I
10.1007/s00521-021-06543-0
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper presents a novel probabilistic distributed framework based on movement primitives for flexible robot assembly. Since the modern advanced industrial cell usually deals with various scenarios that are not fixed via-point trajectories but highly reconfigurable tasks, the industrial robots used in these applications must be capable of adapting and learning new in-demand skills without programming experts. Therefore, we propose a probabilistic framework that could accommodate various learning abilities trained with different movement-primitive datasets, separately. Derived from the Bayesian Committee Machine, this framework could infer new adapting trajectories with weighted contributions of each training dataset. To verify the feasibility of our proposed imitation learning framework, the simulation comparison with the state-of-the-art movement learning framework task-parametrised GMM is conducted. Several key aspects, such as generalisation capability, learning accuracy and computation expense, are discussed and compared. Moreover, two real-world experiments, i.e. riveting picking and nutplate picking, are further tested with the YuMi collaborative robot to verify the application feasibility in industrial assembly manufacturing.
引用
收藏
页码:23453 / 23464
页数:12
相关论文
共 50 条
  • [41] An adaptive learning and control framework based on dynamic movement primitives with application to human-robot handovers
    Wu, Min
    Taetz, Bertram
    He, Yanhao
    Bleser, Gabriele
    Liu, Steven
    ROBOTICS AND AUTONOMOUS SYSTEMS, 2022, 148
  • [42] A Dynamic Movement Primitives-Based Tool Use Skill Learning and Transfer Framework for Robot Manipulation
    Lu, Zhenyu
    Wang, Ning
    Yang, Chenguang
    IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, 2025, 22 : 1748 - 1763
  • [43] A Dynamic Movement Primitives-Based Tool Use Skill Learning and Transfer Framework for Robot Manipulation
    Lu, Zhenyu
    Wang, Ning
    Yang, Chenguang
    IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, 2025, 22 : 1748 - 1763
  • [44] A Unified, Flexible Framework in Network Topology Generation for Distributed Machine Learning
    Liu, Jianhao
    Li, Xiao-Yan
    Liu, Yanhua
    Fan, Weibei
    PROCEEDINGS OF THE 7TH ASIA-PACIFIC WORKSHOP ON NETWORKING, APNET 2023, 2023, : 196 - 197
  • [45] Retrosynthesis prediction with an interpretable deep-learning framework based on molecular assembly tasks
    Wang, Yu
    Pang, Chao
    Wang, Yuzhe
    Jin, Junru
    Zhang, Jingjie
    Zeng, Xiangxiang
    Su, Ran
    Zou, Quan
    Wei, Leyi
    NATURE COMMUNICATIONS, 2023, 14 (01)
  • [46] Retrosynthesis prediction with an interpretable deep-learning framework based on molecular assembly tasks
    Yu Wang
    Chao Pang
    Yuzhe Wang
    Junru Jin
    Jingjie Zhang
    Xiangxiang Zeng
    Ran Su
    Quan Zou
    Leyi Wei
    Nature Communications, 14 (1)
  • [47] A Novel Dynamic Movement Primitives-based Skill Learning and Transfer Framework for Multi-Tool Use
    Lu, Zhenyu
    Wang, Ning
    Li, Miao
    Yang, Chenguang
    2022 IEEE 17TH INTERNATIONAL CONFERENCE ON CONTROL & AUTOMATION, ICCA, 2022, : 1 - 8
  • [48] Toward enhancing the distributed video coder under a multiview video codec framework
    Lee, Shih-Chieh
    Chen, Jiann-Jone
    Tsai, Yao-Hong
    Chen, Chin-Hua
    JOURNAL OF ELECTRONIC IMAGING, 2016, 25 (06)
  • [49] Physiological workload assessment for highly flexible fine-motory assembly tasks using machine learning
    Brillinger, Markus
    Manfredi, Samuel
    Leder, Dominik
    Bloder, Martin
    Jaeger, Markus
    Diwold, Konrad
    Kajmakovic, Amer
    Haslgruebler, Michael
    Pichler, Rudolf
    Brunner, Martin
    Mehr, Stefan
    Malisa, Viktorijo
    COMPUTERS & INDUSTRIAL ENGINEERING, 2024, 188
  • [50] Flexible assembly skill self-learning of robot under multiple constraints
    Song R.
    Li F.-M.
    Quan W.
    Li Y.-B.
    Kongzhi yu Juece/Control and Decision, 2022, 37 (05): : 1329 - 1337