Self-Learning Optimal Control of Nonlinear Systems by Qinglai Wei Ruizhuo Song Benkai Li & Xiaofeng Lin

Author:Qinglai Wei, Ruizhuo Song, Benkai Li & Xiaofeng Lin
Language: eng
Format: epub
Publisher: Springer Singapore, Singapore

From Figs. 4.10–4.12, we can see that after 45 iterations, the iterative value function converges to the optimal one. For the policy iteration-based deterministic Q-learning algorithm, the iterative Q function converges to its optimal within 25 iterations, while it take 45 iterations for value iteration algorithm. It shows the effectiveness of the developed Q-learning algorithm. More important, from Figs. 4.11 and 4.12, we can see that the stability property of system (4.47) cannot be guaranteed under the iterative control law by value iteration algorithm. On the other hand, from Figs. 4.7 and 4.8, we can see that system (4.47) is stable under any of the iterative control law by the policy iteration-based deterministic Q-learning algorithm. Therefore, according to the simulation comparisons, the effectiveness of the developed policy iteration-based deterministic Q-learning algorithm can be justified.

Fig. 4.11The iterative state trajectories (From [38] Fig. 4b)

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.