Local and soft feature selection for value function approximation in batch reinforcement learning for robot navigation

被引:0
|
作者
Fatemeh Fathinezhad
Peyman Adibi
Bijan Shoushtarian
Jocelyn Chanussot
机构
[1] Faculty of Computer Engineering,Artificial Intelligence Department
[2] University of Grenoble Alpes,undefined
[3] CNRS,undefined
[4] Grenoble INP,undefined
[5] GIPSA-lab,undefined
来源
The Journal of Supercomputing | 2024年 / 80卷
关键词
Reinforcement learning; Value function approximation; Local relevance feature selection; Robot navigation;
D O I
暂无
中图分类号
学科分类号
摘要
This paper proposes a novel method for robot navigation in high-dimensional environments that reduce the dimension of the state space using local and soft feature selection. The algorithm selects relevant features based on local correlations between states, avoiding duplicate inappropriate information and adjusting sensor values accordingly. By optimizing the value function approximation based on the local weighted features of states in the reinforcement learning process, the method shows improvements in the robot’s motion flexibility, learning time, the distance traveled to reach its goal, and the minimization of collisions with obstacles. This approach was tested on an E-puck robot using the Webots robot simulator in different test environments.
引用
收藏
页码:10720 / 10745
页数:25
相关论文
共 50 条
  • [41] Global versus local constructive function approximation for on-line reinforcement learning
    Vamplew, P
    Ollington, R
    AI 2005: ADVANCES IN ARTIFICIAL INTELLIGENCE, 2005, 3809 : 113 - 122
  • [42] Practical Bayesian Inverse Reinforcement Learning for Robot Navigation
    Okal, Billy
    Arras, Kai O.
    MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2016, PT III, 2016, 9853 : 271 - 274
  • [43] Mobile Robot Navigation based on Deep Reinforcement Learning
    Ruan, Xiaogang
    Ren, Dingqi
    Zhu, Xiaoqing
    Huang, Jing
    PROCEEDINGS OF THE 2019 31ST CHINESE CONTROL AND DECISION CONFERENCE (CCDC 2019), 2019, : 6174 - 6178
  • [44] Robot Navigation in Crowded Environments: A Reinforcement Learning Approach
    Caruso, Matteo
    Regolin, Enrico
    Verdu, Federico Julian Camerota
    Russo, Stefano Alberto
    Bortolussi, Luca
    Seriani, Stefano
    MACHINES, 2023, 11 (02)
  • [45] Nonholonomic Robot Navigation of Mazes using Reinforcement Learning
    Gleason, Daniel
    Jenkin, Michael
    PROCEEDINGS OF THE 19TH INTERNATIONAL CONFERENCE ON INFORMATICS IN CONTROL, AUTOMATION AND ROBOTICS (ICINCO), 2022, : 369 - 376
  • [46] An Analysis of Feature Selection and Reward Function for Model-Based Reinforcement Learning
    Shen, Shitian
    Lin, Chen
    Mostafavi, Behrooz
    Barnes, Tiffany
    Chi, Min
    INTELLIGENT TUTORING SYSTEMS, ITS 2016, 2016, 9684 : 504 - 505
  • [47] Reinforcement learning-based mobile robot navigation
    Altuntas, Nihal
    Imal, Erkan
    Emanet, Nahit
    Ozturk, Ceyda Nur
    TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, 2016, 24 (03) : 1747 - 1767
  • [48] Fuzzy reinforcement learning and its application in robot navigation
    Duan, Y
    Xu, XH
    Proceedings of 2005 International Conference on Machine Learning and Cybernetics, Vols 1-9, 2005, : 899 - 904
  • [49] Reinforcement Learning Based Approach For Mobile Robot Navigation
    Jaseem, Mohammed M.
    Mathew, Robins
    Hiremath, Somashekhar S.
    PROCEEDINGS OF 2019 INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND KNOWLEDGE ECONOMY (ICCIKE' 2019), 2019, : 524 - 527
  • [50] Automated Feature Selection: A Reinforcement Learning Perspective
    Liu, Kunpeng
    Fu, Yanjie
    Wu, Le
    Li, Xiaolin
    Aggarwal, Charu
    Xiong, Hui
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2023, 35 (03) : 2272 - 2284