Composite Learning Robot Control With Friction Compensation: A Neural Network-Based Approach

被引:115
|
作者
Guo, Kai [1 ,2 ]
Pan, Yongping [3 ]
Yu, Haoyong [3 ]
机构
[1] Shandong Univ, Natl Demonstrat Ctr Expt Mech Engn Educ, Dept Mech Engn, Minist Educ,Key Lab High Efficiency & Clean Mech, Jinan 250061, Shandong, Peoples R China
[2] Zhejiang Univ, State Key Lab Fluid Power & Mechtron Syst, Hangzhou 310027, Peoples R China
[3] Natl Univ Singapore, Dept Biomed Engn, Singapore 117583, Singapore
基金
中国博士后科学基金; 中国国家自然科学基金;
关键词
Adaptive control; composite learning; friction compensation; neural network (NN) approximation; robot manipulator; ADAPTIVE-CONTROL; PARAMETER-ESTIMATION; MANIPULATORS; DESIGN; FEEDFORWARD;
D O I
10.1109/TIE.2018.2886763
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Friction is one of the significant obstacles that hinders high-performance robot tracking control because accurate friction modeling and effective compensation are challenging issues. To address this problem, in this paper, we propose a modified neural network (NN) structure with additional jump approximation activation functions to model the inherent discontinuous friction in robotic systems, this structure allows us to improve the NN approximation accuracy without using too many NN nodes. The modeling accuracy is theoretically guaranteed by a composite learning technique, it explores both online historical data and instantaneous data to achieve NN weight convergence under a much weaker interval-excitation condition than the stringent persistent-excitation condition. Furthermore, a partitioned NN technique is used to handle a problem caused by variable substitution when formulating the prediction error for composite learning. This technique also helps us to alleviate the requirements regarding the inertial matrix inversion and joint acceleration signals. The practical exponential stability of the closed-loop system is proved under the more realizable interval-excitation condition. Experimental results demonstrate the effectiveness and superiority of the proposed approach.
引用
收藏
页码:7841 / 7851
页数:11
相关论文
共 50 条
  • [31] Adaptive FIT-SMC Approach for an Anthropomorphic Manipulator With Robust Exact Differentiator and Neural Network-Based Friction Compensation
    Ali, Khurram
    Ullah, Safeer
    Mehmood, Adeel
    Mostafa, Hala
    Marey, Mohamed
    Iqbal, Jamshed
    IEEE ACCESS, 2022, 10 : 3378 - 3389
  • [32] BIOINSPIRED NEURAL NETWORK-BASED Q-LEARNING APPROACH FOR ROBOT PATH PLANNING IN UNKNOWN ENVIRONMENTS
    Ni, Jianjun
    Li, Xinyun
    Hua, Mingang
    Yang, Simon X.
    INTERNATIONAL JOURNAL OF ROBOTICS & AUTOMATION, 2016, 31 (06): : 464 - 474
  • [33] Friction compensation of a retrofit milling machine through a fuzzy logic controller combined with neural network-based feedforward control
    Yan, MT
    Huang, YL
    Hsue, AWJ
    2005 IEEE INTERNATIONAL CONFERENCE ON MECHATRONICS, 2005, : 398 - 403
  • [34] Robot inverse dynamics control with feedback and compensation based on neural network
    Tan, YM
    Mao, ZY
    1997 IEEE INTERNATIONAL CONFERENCE ON INTELLIGENT PROCESSING SYSTEMS, VOLS 1 & 2, 1997, : 1297 - 1301
  • [35] A neural network-based approach for the control of autoleveller in draw frame
    Su, QX
    Deng, N
    QUALITY TEXTILES FOR QUALITY LIFE, VOLS 1-4, 2004, : 1355 - 1356
  • [36] A neural network-based approach to hybrid systems identification for control
    Fabiani, Filippo
    Stellato, Bartolomeo
    Masti, Daniele
    Goulart, Paul J.
    AUTOMATICA, 2025, 174
  • [37] Neural network-based compensation control of mobile robots with partially known structure
    Rossomando, F. G.
    Soria, C.
    Carelli, R.
    IET CONTROL THEORY AND APPLICATIONS, 2012, 6 (12): : 1851 - 1860
  • [38] Robust adaptive neural network-based compensation control of a class of quadrotor aircrafts
    Jin, Xiao-Zheng
    He, Tao
    Wu, Xiao-Ming
    Wang, Hai
    Chi, Jing
    JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS, 2020, 357 (17): : 12241 - 12263
  • [39] Neural Network-Based Repetitive Learning Control of Euler Lagrange Systems: An Output Feedback Approach
    Tatlicioglu, Enver
    Cobanoglu, Necati
    Zergeroglu, Erkan
    IEEE CONTROL SYSTEMS LETTERS, 2018, 2 (01): : 13 - 18
  • [40] Artificial Neural Network-based Approach for Inverse Kinematics of PUMA 260 Robot
    Shareef, Israa Rafie
    INTERNATIONAL JOURNAL OF INTEGRATED ENGINEERING, 2024, 16 (05): : 373 - 384