Neuro-dynamic programming with recurrent critic for automatic train regulation of metro line.

Wei-Song Lin Jih-Wen Sheu

Published in: IJCNN (2009)

Keyphrases

dynamic programming
state space
neural network
artificial neural networks
fully automatic
line segments
semi automatic
approximate dynamic programming
real time
infinite horizon
optimal control
recurrent neural networks
optimal policy
data driven
hidden markov models
website
information systems
learning algorithm