Fuzzy Reinforcement Learning and Curriculum Transfer Learning for Micromanagement in Multi-Robot Confrontation

被引:4
作者
Hu, Chunyang [1 ]
Xu, Meng [2 ]
机构
[1] Hubei Univ Arts & Sci, Sch Comp Engn, Xiangyang 441053, Peoples R China
[2] Northwestern Polytech Univ, Sch Comp Sci, Xian 710072, Shaanxi, Peoples R China
关键词
multi-robot confrontation; fuzzy reinforcement learning; curriculum transfer learning; neural network; INTELLIGENCE; FRAMEWORK;
D O I
10.3390/info10110341
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Multi-Robot Confrontation on physics-based simulators is a complex and time-consuming task, but simulators are required to evaluate the performance of the advanced algorithms. Recently, a few advanced algorithms have been able to produce considerably complex levels in the context of the robot confrontation system when the agents are facing multiple opponents. Meanwhile, the current confrontation decision-making system suffers from difficulties in optimization and generalization. In this paper, a fuzzy reinforcement learning (RL) and the curriculum transfer learning are applied to the micromanagement for robot confrontation system. Firstly, an improved Q-learning in the semi-Markov decision-making process is designed to train the agent and an efficient RL model is defined to avoid the curse of dimensionality. Secondly, a multi-agent RL algorithm with parameter sharing is proposed to train the agents. We use a neural network with adaptive momentum acceleration as a function approximator to estimate the state-action function. Then, a method of fuzzy logic is used to regulate the learning rate of RL. Thirdly, a curriculum transfer learning method is used to extend the RL model to more difficult scenarios, which ensures the generalization of the decision-making system. The experimental results show that the proposed method is effective.
引用
收藏
页数:22
相关论文
共 50 条
[31]   Multi-agent reinforcement learning with approximate model learning for competitive games [J].
Park, Young Joon ;
Cho, Yoon Sang ;
Kim, Seoung Bum .
PLOS ONE, 2019, 14 (09)
[32]   A multi-task transfer learning method with dictionary learning [J].
Zheng, Xin ;
Lin, Luyue ;
Liu, Bo ;
Xiao, Yanshan ;
Xiong, Xiaoming .
KNOWLEDGE-BASED SYSTEMS, 2020, 191
[33]   Reinforcement Learning based on Modular Fuzzy Model with Gating Unit [J].
Watanabe, Toshihiko ;
Wada, Tatsuya .
2008 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC), VOLS 1-6, 2008, :1805-+
[34]   A fuzzy reinforcement learning algorithm for inventory control in supply chains [J].
Mohammad Hossein Fazel Zarandi ;
Seyed Vahid Moosavi ;
Marzieh Zarinbal .
The International Journal of Advanced Manufacturing Technology, 2013, 65 :557-569
[35]   Fuzzy reinforcement learning control for compliance tasks of robotic manipulators [J].
Tzafestas, SG ;
Rigatos, GG .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2002, 32 (01) :107-113
[36]   A fuzzy reinforcement learning algorithm for inventory control in supply chains [J].
Zarandi, Mohammad Hossein Fazel ;
Moosavi, Seyed Vahid ;
Zarinbal, Marzieh .
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY, 2013, 65 (1-4) :557-569
[37]   Type-2 Fuzzy Formation Control for Collision-Free Multi-Robot Systems [J].
Chang, Yeong-Hwa ;
Chen, Chun-Lin ;
Chan, Wei-Shou ;
Lin, Hung-Wei .
INTERNATIONAL JOURNAL OF FUZZY SYSTEMS, 2013, 15 (04) :435-451
[38]   Adaptive multi-model fusion learning for sparse-reward reinforcement learning [J].
Park, Giseung ;
Jung, Whiyoung ;
Han, Seungyul ;
Choi, Sungho ;
Sung, Youngchul .
NEUROCOMPUTING, 2025, 633
[39]   Table-Balancing Cooperative Robot Based on Deep Reinforcement Learning [J].
Kim, Yewon ;
Kim, Dae-Won ;
Kang, Bo-Yeong .
SENSORS, 2023, 23 (11)
[40]   Reinforcement Learning-Based Tracking Control For Wheeled Mobile Robot [J].
Nguyen Tan Luy .
PROCEEDINGS 2012 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC), 2012, :462-467