Self-Learning Optimal Control for Uncertain Nonlinear Systems via Online Updated Cost Function

被引:0
作者
Zhao, Bo [1 ]
Shi, Guang [2 ]
Li, Chao [2 ]
机构
[1] Chinese Acad Sci, State Key Lab Management & Control Complex Syst, Inst Automat, Beijing, Peoples R China
[2] Coordinat Ctr China, Natl Comp Network Emergency Response Tech Team, Beijing, Peoples R China
来源
PROCEEDINGS 2018 33RD YOUTH ACADEMIC ANNUAL CONFERENCE OF CHINESE ASSOCIATION OF AUTOMATION (YAC) | 2018年
基金
北京市自然科学基金; 中国国家自然科学基金;
关键词
Adaptive dynamic programming; Uncertain nonlinear systems; Optimal control; Disturbance observer; Neural networks; Reinforcement learning; DISTURBANCE OBSERVER; DESIGN;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper presents an online updated cost function based self-learning optimal control scheme for uncertain nonlinear systems. By establishing an online updated cost function with the help of disturbance observer, the Hamilton-Jacobi-Bellman equation is solved by constructing a critic neural network, whose weight vector is tuned by self-learning algorithm. And then, the optimal control scheme is derived indirectly. Based on Lyapunov stability analysis, the closed-loop system with the proposed scheme is guaranteed to be stable. The simulation results show the effectiveness of the developed self-learning optimal control scheme. The cost function reflects the system uncertainties in real time, which implies that this method relaxes the assumptions on available upper-bounds and matching condition for system dynamics in compared with many existing methods.
引用
收藏
页码:1061 / 1065
页数:5
相关论文
共 18 条
  • [1] [Anonymous], 1992, HDB INTELLIGENT CONT
  • [2] A novel actor-critic-identifier architecture for approximate optimal control of uncertain nonlinear systems
    Bhasin, S.
    Kamalapurkar, R.
    Johnson, M.
    Vamvoudakis, K. G.
    Lewis, F. L.
    Dixon, W. E.
    [J]. AUTOMATICA, 2013, 49 (01) : 82 - 92
  • [3] Robust Adaptive Dynamic Programming and Feedback Stabilization of Nonlinear Systems
    Jiang, Yu
    Jiang, Zhong-Ping
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2014, 25 (05) : 882 - 893
  • [4] A self-learning disturbance observer for nonlinear systems in feedback-error learning scheme
    Kayacan, Erkan
    Peschel, Joshua M.
    Chowdhary, Girish
    [J]. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2017, 62 : 276 - 285
  • [5] Reinforcement Learning and Feedback Control USING NATURAL DECISION METHODS TO DESIGN OPTIMAL ADAPTIVE CONTROLLERS
    Lewis, Frank L.
    Vrabie, Draguna
    Vamvoudakis, Kyriakos G.
    [J]. IEEE CONTROL SYSTEMS MAGAZINE, 2012, 32 (06): : 76 - 105
  • [6] Liu D, 2017, ADV IND CONTROL, P1, DOI 10.1007/978-3-319-50815-3
  • [7] Adaptive Optimal Control Using Frequency Selective Information of the System Uncertainty With Application to Unmanned Aircraft
    Maity, Arnab
    Hoecht, Leonhard
    Heise, Christian
    Holzapfel, Florian
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2018, 48 (01) : 165 - 177
  • [8] Off-Policy Actor-Critic Structure for Optimal Control of Unknown Systems With Disturbances
    Song, Ruizhuo
    Lewis, Frank L.
    Wei, Qinglai
    Zhang, Huaguang
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2016, 46 (05) : 1041 - 1050
  • [9] Disturbance observer-based robust missile autopilot design with full-state constraints via adaptive dynamic programming
    Sun, Jingliang
    Liu, Chunsheng
    [J]. JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS, 2018, 355 (05): : 2344 - 2368
  • [10] Optimal Robust Linear Quadratic Regulator for Systems Subject to Uncertainties
    Terra, Marco H.
    Cerri, Joao P.
    Ishihara, Joao Y.
    [J]. IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2014, 59 (09) : 2586 - 2591