Journal
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
Volume 67, Issue 7, Pages 3582-3589Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TAC.2021.3108552
Keywords
Heuristic algorithms; Trajectory; Approximation algorithms; Stochastic processes; Dynamic programming; Data models; Computational modeling; Reinforcement learning; stochastic control; nonlinear systems
Funding
- National Science Foundation [ECCS-1637889, CDSE1802867]
- AFOSR DDIP Grant [FA9550-17-1-0068]
- NSFC [61801213]
- National Science Foundation (NSF) [CRII-CPS-1850206]
Ask authors/readers for more resources
This article addresses the problem of learning the optimal control policy for a nonlinear stochastic dynamical system and proposes a decoupled data-based control (D2C) algorithm to solve it. Experimental results show that the performance of this algorithm is nearly optimal, and the training time is significantly reduced compared to other algorithms.
This article addresses the problem of learning the optimal control policy for a nonlinear stochastic dynamical. This problem is subject to the curse of dimensionality associated with the dynamic programming method. This article proposes a novel decoupled data-based control (D2C) algorithm that addresses this problem using a decoupled, open-loop-closed-loop, approach. First, an open-loop deterministic trajectory optimization problem is solved using a black-box simulation model of the dynamical system. Then, closed-loop control is developed around this open-loop trajectory by linearization of the dynamics about this nominal trajectory. By virtue of linearization, a linear quadratic regulator based algorithm can be used for this closed-loop control. We show that the performance of D2C algorithm is approximately optimal. Moreover, simulation performance suggests a significant reduction in training time compared to other state-of-the-art algorithms.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available