Automated conceptual design of mechanisms based on Thompson Sampling and Monte Carlo Tree Search

被引:0
作者
Mao, Jiangmin [1 ,2 ]
Zhu, Yingdan [1 ,2 ]
Chen, Gang [1 ]
Yan, Chun [1 ]
Zhang, Wuxiang [3 ]
机构
[1] Chinese Acad Sci, Ningbo Inst Mat Technol & Engn, Zhejiang Prov Key Lab Robot & Intelligent Mfg Equi, Ningbo 315201, Peoples R China
[2] Univ Chinese Acad Sci, Ctr Mat Sci & Optoelect Engn, Beijing 100049, Peoples R China
[3] Beihang Univ, Sch Mech Engn & Automat, Beijing 100191, Peoples R China
关键词
Mechanism synthesis; Conceptual design; Monte Carlo Tree Search; Thompson Sampling; Dirichlet distribution; FUNCTIONAL SYNTHESIS; ENUMERATION; ALGORITHM; FRAMEWORK;
D O I
10.1016/j.asoc.2024.112659
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Conceptual design of mechanisms is a crucial part of achieving product innovation as mechanisms perform the transmission and transformation of specific motions in the machine. However, existing approaches for automated synthesis of mechanisms are either inefficient or prone to a loss of optimal solutions. To fill this gap, a systematic online decision-making method using Thompson Sampling (TS) based Monte Carlo Tree Search (MCTS) for automated conceptual design of mechanisms is proposed. The functional transformation relationships between inputs and outputs of the intended mechanism system are used to determine combinatorial patterns. Then, a functional representation model is constructed based on the combination rules of motion features and the inference relationships of function elements to represent a range of primitive mechanisms as fundamental building blocks. Finally, the optimal action selection strategy based on TS is applied into MCTS to develop Dirichlet based Monte Carlo Tree Search (D-MCTS) algorithm for searching mechanism building blocks. In addition, the conceptual design of the beat-up mechanism as well as the stitching and feeding mechanism are conducted to validate the feasibility of the proposed approach. Compared with specialized heuristics, D-MCTS achieves higher efficiency in finding the best combination of mechanism building blocks. Compared with other common algorithms, D-MCTS can always avoid the local optima trap to find the global optimal solution without any necessary hyper-parameter tuning. The proposed method exhibits a more balanced performance in explo- ration and exploitation, which provides better solutions for mechanism synthesis of given requirements.
引用
收藏
页数:23
相关论文
共 67 条
  • [1] LASER: A Scalable Response Prediction Platform For Online Advertising
    Agarwal, Deepak
    Long, Bo
    Traupman, Jonathan
    Xin, Doris
    Zhang, Liang
    [J]. WSDM'14: PROCEEDINGS OF THE 7TH ACM INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING, 2014, : 173 - 182
  • [2] A novice-centered decision-support system for type synthesis of function-generation mechanisms
    Al-Dwairi, Abdullah F.
    Dweiri, Fikri T.
    Ashour, Omar M.
    [J]. MECHANISM AND MACHINE THEORY, 2010, 45 (09) : 1252 - 1268
  • [3] A dynamic thompson sampling hyper-heuristic framework for learning activity planning in personalized learning
    Aslan, Ayse
    Bakir, Ilke
    Vis, Iris F. A.
    [J]. EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2020, 286 (02) : 673 - 688
  • [4] Bai A., 2013, Adv. Neural Inf. Process. Syst., P1
  • [5] Posterior sampling for Monte Carlo planning under uncertainty
    Bai, Aijun
    Wu, Feng
    Chen, Xiaoping
    [J]. APPLIED INTELLIGENCE, 2018, 48 (12) : 4998 - 5018
  • [6] Baudry D, 2021, ADV NEUR IN, V34
  • [7] Flexible slider crank mechanism synthesis using meta-heuristic optimization techniques: a new designer tool assistance for a compliant mechanism synthesis
    Ben Abdallah, Mohamed Amine
    Khemili, Imed
    Aifaoui, Nizar
    [J]. ARTIFICIAL INTELLIGENCE REVIEW, 2020, 53 (04) : 2809 - 2840
  • [8] Three- dimensional circular various weave patterns in woven preform structures
    Bilisik, Kadir
    Karaduman, Nesrin Sahbaz
    Bilisik, Nedim Erman
    Bilisik, Havva Esra
    [J]. TEXTILE RESEARCH JOURNAL, 2014, 84 (06) : 638 - 654
  • [9] A Survey of Monte Carlo Tree Search Methods
    Browne, Cameron B.
    Powley, Edward
    Whitehouse, Daniel
    Lucas, Simon M.
    Cowling, Peter I.
    Rohlfshagen, Philipp
    Tavener, Stephen
    Perez, Diego
    Samothrakis, Spyridon
    Colton, Simon
    [J]. IEEE TRANSACTIONS ON COMPUTATIONAL INTELLIGENCE AND AI IN GAMES, 2012, 4 (01) : 1 - 43
  • [10] An evolutionary algorithm for path synthesis of mechanisms
    Cabrera, J. A.
    Ortiz, A.
    Nadal, F.
    Castillo, J. J.
    [J]. MECHANISM AND MACHINE THEORY, 2011, 46 (02) : 127 - 141