Optimal and Autonomous Control Using Reinforcement Learning: A Survey
Optimal and Autonomous Control Using Reinforcement Learning: A Survey
复制标题
使用强化学习的最优自主控制：调查
DOI：
10.1109/tnnls.2017.2773458
复制
发表时间：
2018-06-01
影响因子：
10.4
通讯作者：
Lewis, Frank L.
中科院分区：
文献类型：
--
作者：
Kiumarsi, Bahare;Vamvoudakis, Kyriakos G.;Lewis, Frank L.
本文综述了基于强化学习（RL）的反馈控制解决方案的现状，以实现单智能体和多智能体系统的最优调节和跟踪。现有的RL解决方案的最优H-2和h -∞控制问题，以及图形游戏，将进行审查。RL方法在线学习最优控制和博弈问题的解决方案，并使用沿系统轨迹的测量数据。我们分别讨论了q -学习和积分RL算法作为离散时间（DT）和连续时间（CT）系统的核心算法。此外，我们还讨论了CT和DT系统的非策略RL的新方向。最后，我们回顾了几个应用。
This paper reviews the current state of the art on reinforcement learning (RL)-based feedback control solutions to optimal regulation and tracking of single and multiagent systems. Existing RL solutions to both optimal H-2 and H-infinity control problems, as well as graphical games, will be reviewed. RL methods learn the solution to optimal control and game problems online and using measured data along the system trajectories. We discuss Q-learning and the integral RL algorithm as core algorithms for discrete-time (DT) and continuous-time (CT) systems, respectively. Moreover, we discuss a new direction of off-policy RL for both CT and DT systems. Finally, we review several applications.