TY - JOUR
T1 - Reinforcement Learning-Based Nearly Optimal Control for Constrained-Input Partially Unknown Systems Using Differentiator
AU - Guo, Xinxin
AU - Yan, Weisheng
AU - Cui, Rongxin
N1 - Publisher Copyright:
© 2012 IEEE.
PY - 2020/11
Y1 - 2020/11
N2 - In this article, a synchronous reinforcement-learning-based algorithm is developed for input-constrained partially unknown systems. The proposed control also alleviates the need for an initial stabilizing control. A first-order robust exact differentiator is employed to approximate unknown drift dynamics. Critic, actor, and disturbance neural networks (NNs) are established to approximate the value function, the control policy, and the disturbance policy, respectively. The Hamilton-Jacobi-Isaacs equation is solved by applying the value function approximation technique. The stability of the closed-loop system can be ensured. The state and weight errors of the three NNs are all uniformly ultimately bounded. Finally, the simulation results are provided to verify the effectiveness of the proposed method.
AB - In this article, a synchronous reinforcement-learning-based algorithm is developed for input-constrained partially unknown systems. The proposed control also alleviates the need for an initial stabilizing control. A first-order robust exact differentiator is employed to approximate unknown drift dynamics. Critic, actor, and disturbance neural networks (NNs) are established to approximate the value function, the control policy, and the disturbance policy, respectively. The Hamilton-Jacobi-Isaacs equation is solved by applying the value function approximation technique. The stability of the closed-loop system can be ensured. The state and weight errors of the three NNs are all uniformly ultimately bounded. Finally, the simulation results are provided to verify the effectiveness of the proposed method.
KW - First-order robust exact differentiator (RED)
KW - input constraint
KW - neural network (NN)
KW - reinforcement learning (RL)
KW - two-player zero-sum game
UR - http://www.scopus.com/inward/record.url?scp=85077270909&partnerID=8YFLogxK
U2 - 10.1109/TNNLS.2019.2957287
DO - 10.1109/TNNLS.2019.2957287
M3 - 文章
C2 - 31880567
AN - SCOPUS:85077270909
SN - 2162-237X
VL - 31
SP - 4713
EP - 4725
JO - IEEE Transactions on Neural Networks and Learning Systems
JF - IEEE Transactions on Neural Networks and Learning Systems
IS - 11
M1 - 8943132
ER -