Reinforcement Learning for Linear Continuous-time Systems: an Incremental Learning Approach

被引：20

作者：

Bian, Tao ^{[1
]}

Jiang, Zhong-Ping ^{[2
]}

机构：

[1] Bank Amer Merrill Lynch, One Bryant Pk, New York, NY 10036 USA

[2] NYU, Tandon Sch Engn, Dept Elect & Comp Engn, Control & Networks Lab, 5 Metrotech Ctr, Brooklyn, NY 11201 USA

来源：

IEEE-CAA JOURNAL OF AUTOMATICA SINICA | 2019年 / 6卷 / 02期

基金：

美国国家科学基金会;

关键词：

Adaptive optimal control; robust dynamic programming; value iteration (VI); ADAPTIVE OPTIMAL-CONTROL; STABILIZATION; STATE;

D O I：

10.1109/JAS.2019.1911390

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper, we introduce a novel reinforcement learning (RL) scheme for linear continuous-time dynamical systems. Different from traditional batch learning algorithms, an incremental learning approach is developed, which provides a more efficient way to tackle the on-line learning problem in real-world applications. We provide concrete convergence and robust analysis on this incremental-learning algorithm. An extension to solving robust optimal control problems is also given. Two simulation examples are also given to illustrate the effectiveness of our theoretical result.

引用

页码：433 / 440

页数：8

共 50 条

[21] Event-Triggered Reinforcement Learning Approach for Unknown Nonlinear Continuous-Time System
Zhong, Xiangnan
Ni, Zhen
He, Haibo
Xu, Xin
Zhao, Dongbin
PROCEEDINGS OF THE 2014 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2014, : 3677 - 3684
[22] Traffic-signal control reinforcement learning approach for continuous-time Markov games
Aragon-Gomez, Roman
Clempner, Julio B.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2020, 89
[23] Value iteration based integral reinforcement learning approach for H∞ controller design of continuous-time nonlinear systems
Xiao, Geyang
Zhang, Huaguang
Zhang, Kun
Wen, Yinlei
NEUROCOMPUTING, 2018, 285 : 51 - 59
[24] Iterative learning control for linear continuous-time switched systems with observation noise
Yang, Xuan
Ruan, Xiaoe
TRANSACTIONS OF THE INSTITUTE OF MEASUREMENT AND CONTROL, 2019, 41 (04) : 1178 - 1185
[25] Efficient continuous-time reinforcement learning with adaptive state graphs
Neumann, Gerhard
Pfeiffer, Michael
Maass, Wolfgang
MACHINE LEARNING: ECML 2007, PROCEEDINGS, 2007, 4701 : 250 - +
[26] IMPACT OF COMPUTATION IN INTEGRAL REINFORCEMENT LEARNING FOR CONTINUOUS-TIME CONTROL
Cao, Wenhan
Pan, Wei
12th International Conference on Learning Representations, ICLR 2024, 2024,
[27] Integral reinforcement learning-based optimal output feedback control for linear continuous-time systems with input delay
Wang, Gao
Luo, Biao
Xue, Shan
NEUROCOMPUTING, 2021, 460 : 31 - 38
[28] Online reinforcement learning multiplayer non-zero sum games of continuous-time Markov jump linear systems
Xin, Xilin
Tu, Yidong
Stojanovic, Vladimir
Wang, Hai
Shi, Kaibo
He, Shuping
Pan, Tianhong
APPLIED MATHEMATICS AND COMPUTATION, 2022, 412
[29] Linear inverse reinforcement learning in continuous time and space
Kamalapurkar, Rushikesh
2018 ANNUAL AMERICAN CONTROL CONFERENCE (ACC), 2018, : 1683 - 1688
[30] A new approach to the analysis of linear continuous-time systems
Kaczorek, Tadeusz
2023 27TH INTERNATIONAL CONFERENCE ON METHODS AND MODELS IN AUTOMATION AND ROBOTICS, MMAR, 2023, : 1 - 4

← 1 2 3 4 5 →