Robust Learning-Based Control via Bootstrapped Multiplicative Noise

被引:0
|
作者
Gravell, Benjamin [1 ]
Summers, Tyler [1 ]
机构
[1] Univ Texas Dallas, Richardson, TX 75083 USA
关键词
Robust adaptive control; bootstrap; multiplicative noise; regret; SYSTEMS;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Despite decades of research and recent progress in adaptive control and reinforcement learning, there remains a fundamental lack of understanding in designing controllers that provide robustness to inherent non-asymptotic uncertainties arising from models estimated with finite, noisy data. We propose a robust adaptive control algorithm that explicitly incorporates such non-asymptotic uncertainties into the control design. The algorithm has three components: (1) a least-squares nominal model estimator; (2) a bootstrap resampling method that quantifies non-asymptotic variance of the nominal model estimate; and (3) a non-conventional robust control design method using an optimal linear quadratic regulator (LQR) with multiplicative noise. A key advantage of the proposed approach is that the system identification and robust control design procedures both use stochastic uncertainty representations, so that the actual inherent statistical estimation uncertainty directly aligns with the uncertainty the robust controller is being designed against. We show through numerical experiments that the proposed robust adaptive controller can significantly outperform the certainty equivalent controller on both expected regret and measures of regret risk.
引用
收藏
页码:599 / 607
页数:9
相关论文
共 50 条
  • [1] Robust Data-Driven Output Feedback Control via Bootstrapped Multiplicative Noise
    Gravell, Benjamin
    Shames, Iman
    Summers, Tyler
    LEARNING FOR DYNAMICS AND CONTROL CONFERENCE, VOL 168, 2022, 168
  • [2] Robust Control Design for Linear Systems via Multiplicative Noise
    Gravell, Benjamin J.
    Esfahani, Peyman Mohajerin
    Summers, Tyler H.
    IFAC PAPERSONLINE, 2020, 53 (02): : 7392 - 7399
  • [3] Robust Reinforcement Learning for Stochastic Linear Quadratic Control with Multiplicative Noise
    Pang, Bo
    Jiang, Zhong-Ping
    IFAC PAPERSONLINE, 2021, 54 (07): : 240 - 243
  • [4] Provably safe and robust learning-based model predictive control
    Aswani, Anil
    Gonzalez, Humberto
    Sastry, S. Shankar
    Tomlin, Claire
    AUTOMATICA, 2013, 49 (05) : 1216 - 1226
  • [5] Machine learning-based robust trajectory tracking control for FSGR
    Jia, Lin
    Wang, Yaonan
    Zhang, Changfan
    Zhao, Kaihui
    Zhou, Langming
    JOURNAL OF ENGINEERING-JOE, 2019, 2019 (23): : 9220 - 9225
  • [6] Research Progress on Learning-based Robust Adaptive Critic Control
    Wang D.
    Zidonghua Xuebao/Acta Automatica Sinica, 2019, 45 (06): : 1031 - 1043
  • [7] Robust Reinforcement Learning-based Vehicle Control with Object Avoidance
    Lelko, Attila
    Nemeth, Balazs
    Mihaly, Andras
    Sename, Olivier
    Gaspar, Peter
    IFAC PAPERSONLINE, 2024, 58 (10): : 134 - 139
  • [8] Learning-Based Distributionally Robust Motion Control with Gaussian Processes
    Hakobyan, Astghik
    Yang, Insoon
    2020 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2020, : 7667 - 7674
  • [9] Learning-based robust control methodologies under information constraints
    Karimi, Hamid Reza
    Wang, Ning
    Man, Zhihong
    INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL, 2022, 32 (05) : 2467 - 2471
  • [10] Learning-Based Quantum Robust Control: Algorithm, Applications, and Experiments
    Dong, Daoyi
    Xing, Xi
    Ma, Hailan
    Chen, Chunlin
    Liu, Zhixin
    Rabitz, Herschel
    IEEE TRANSACTIONS ON CYBERNETICS, 2020, 50 (08) : 3581 - 3593