Optimal control of multiple CSPS system based on event-based Q learning.
Hao TangHaifeng WanLei ZhouJianghong HanPublished in: CDC (2009)
Keyphrases
- optimal control
- reinforcement learning
- dynamic programming
- control problems
- control strategy
- optimal control problems
- feedback control
- risk sensitive
- infinite horizon
- state space
- function approximation
- class of nonlinear systems
- actor critic
- policy iteration
- learning algorithm
- machine learning
- constraint satisfaction
- constraint satisfaction problems
- brownian motion
- data mining
- markov decision processes
- evolutionary algorithm
- multi agent
- linear quadratic