H∞ optimal control of unknown linear discrete-time systems: An off-policy reinforcement learning approach.
Bahare KiumarsiHamidreza ModaresFrank L. LewisZhong-Ping JiangPublished in: RAM/CIS (2015)
Keyphrases
- optimal control
- reinforcement learning
- linear quadratic
- optimal control problems
- control problems
- dynamic programming
- state space
- control strategy
- infinite horizon
- average cost
- risk sensitive
- class of nonlinear systems
- neural network
- brownian motion
- feedback control
- markov chain
- sufficient conditions
- machine learning