Q-learning for Optimal Control of Continuous-time Systems.
Biao LuoDerong LiuTingwen HuangPublished in: CoRR (2014)
Keyphrases
- optimal control
- reinforcement learning
- dynamic programming
- control problems
- risk sensitive
- feedback control
- optimal control problems
- state space
- class of nonlinear systems
- learning algorithm
- neural network
- actor critic
- policy iteration
- control law
- control strategy
- function approximation
- reinforcement learning methods
- multistage
- markov chain
- data mining
- real time