Ramp metering based on on-line ADHDP (lambda) controller.
Xuerui BaiDongbin ZhaoJianqiang YiJing XuPublished in: IJCNN (2008)
Keyphrases
- actor critic
- optimal control
- reinforcement learning
- policy gradient
- temporal difference
- neuro fuzzy
- approximate dynamic programming
- real time
- gradient method
- control system
- reinforcement learning algorithms
- function approximation
- control method
- fuzzy controller
- control algorithm
- control strategy
- pid controller
- objective function
- policy iteration
- markov chain
- controller design
- average reward
- closed loop
- least squares
- state space